Dev Career Secrets from a GitHub Strategy Leader ๐Ÿ”— **Source:** [Dev Career Secrets from a GitHub Strategy Leader](https://www.youtube.com/watch?v=I9zQos_ePDs) > [!note]+ Document Notes > tags: [developer-careers, tech-hiring, github-profiles] > author: Francesco (The Talk with Francesco) > source: YouTube > reference: https://youtu.be/I9zQos_ePDs > date created: 2024-12-28 10:00 > > # ๐Ÿท๏ธ Title: Dev Career Secrets from a GitHub Strategy Leader > <!-- deep-read:sections start --> ## ๐Ÿงฐ Action Inbox - [H1] Update GitHub profile README with specific technologies and personality โ€” 10 seconds recruiter scan #career - [H2] Pin repositories relevant to target role, not everything โ€” quality over quantity showcase #portfolio - [H3] Document learning journey publicly on GitHub discussions โ€” build community reputation #networking ## ๐Ÿงญ Table of Contents - [๐Ÿ“ Summary](#-summary) - [๐Ÿ“– Main Content](#-main-content) - [๐Ÿ—’๏ธ Notes](#๏ธ-notes) - [๐Ÿ“š References](#-references) - [๐Ÿ—ฃ๏ธ Paraphrase](#๏ธ-paraphrase) - [๐Ÿ“ฐ Newsletter](#-newsletter) - [โœ… Action Items](#-action-items) - [๐Ÿงช Hypotheses](#-hypotheses) - [๐Ÿ“Š Methodology](#-methodology) - [โ“ Questions](#-questions) - [๐Ÿ”ฎ Future Research](#-future-research) - [๐Ÿ’ก Implications](#-implications) - [๐Ÿง  Learning Styles Assessment](#-learning-styles-assessment) - [๐Ÿ“Š Accessibility Assessment](#-accessibility-assessment) - [๐ŸŒ DEIB Assessment](#-deib-assessment) - [โœ… Logical Validity](#-logical-validity) - [๐Ÿ›ก๏ธ DARVO Analysis](#-darvo-analysis) - [๐ŸŽฏ Conclusion](#-conclusion) ### ๐Ÿ”— References - [GitHub State of the Octoverse Report](https://github.com) - [GitHub Discussions](https://github.com) - [Just Another Job Seeking App (JAJA)](https://github.com) ## ๐Ÿ“ Summary - **What happened**: GitHub developer advocate Andrea Griffiths discusses developer career strategies for 2026, addressing layoffs, junior developer challenges, and effective use of AI in job seeking - **Why it matters**: Tech job market has become more competitive with higher barriers to entry, requiring new strategies for career success - **Key stakeholders**: Junior developers, career changers, job seekers in tech - **Risks and tradeoffs**: AI can help learning but over-reliance damages authenticity; networking helps but requires genuine engagement - **Bottom line**: Junior developer roles haven't disappeared but evolved - success requires deeper fundamentals, strategic GitHub presence, and authentic community participation ## ๐Ÿ“– Main Content Andrea Griffiths, a career changer turned GitHub developer advocate, explains how the tech hiring landscape has shifted. Junior developer roles now require broader system thinking beyond basic coding - AI has raised the bar by making simple coding accessible to everyone. Success depends on understanding debugging, security, architecture, and business context. The conversation addresses common misconceptions about GitHub contribution graphs (green squares don't equal hiring potential) and emphasizes strategic profile optimization. Recruiters spend only 10 seconds on initial profile scans, focusing on pinned repositories, bio information, and recent activity quality over quantity. AI should be leveraged as a learning tool and conversation partner, not a replacement for fundamental skills. Companies expect AI usage but value developers who can architect, debug, and make security decisions. Building authentic community presence through documentation, open source contributions, and public learning creates more opportunities than traditional job applications. ## ๐Ÿ—’๏ธ Notes #### โฑ๏ธ [00:00](https://youtu.be/I9zQos_ePDs?t=0) โ€” Introduction and context setting - Key idea: Discussing developer careers for 2026 amid challenging job market - Evidence: "It has been a tough year for most of us layoffs" #### โฑ๏ธ [03:45](https://youtu.be/I9zQos_ePDs?t=225) โ€” Andrea's non-traditional background - Key idea: Career changer perspective on tech industry evolution - Evidence: "I'm a career changer... been doing this now for almost 10 years" #### โฑ๏ธ [06:30](https://youtu.be/I9zQos_ePDs?t=390) โ€” Junior developer role transformation - Key idea: AI has reshaped entry-level expectations, not eliminated roles - Evidence: "No one is born a senior engineer... but now there is a higher bar" #### โฑ๏ธ [09:15](https://youtu.be/I9zQos_ePDs?t=555) โ€” Beyond basic coding skills - Key idea: Modern juniors need architecture and systems thinking from day one - Evidence: "Can you debug that app? Can you put proper security measures in place?" #### โฑ๏ธ [12:00](https://youtu.be/I9zQos_ePDs?t=720) โ€” Green squares misconception - Key idea: GitHub contribution graphs don't indicate developer quality - Evidence: "I don't judge a person's capability by how many green squares" #### โฑ๏ธ [15:20](https://youtu.be/I9zQos_ePDs?t=920) โ€” 10-second recruiter scan strategy - Key idea: First impression depends on pinned repos, bio, and recent activity - Evidence: "Your pin repos... what are you choosing to show me?" #### โฑ๏ธ [18:45](https://youtu.be/I9zQos_ePDs?t=1125) โ€” Making yourself hireable - Key idea: Networking and community participation outweigh cold applications - Evidence: "Nine times out of 10, you're not even really applying for jobs" #### โฑ๏ธ [22:10](https://youtu.be/I9zQos_ePDs?t=1330) โ€” Documentation and storytelling - Key idea: Public learning journeys create opportunities and connections - Evidence: "Humans we love to follow a story... document everything that you're doing" #### โฑ๏ธ [26:30](https://youtu.be/I9zQos_ePDs?t=1590) โ€” Authentic online presence - Key idea: Have opinions and engage genuinely, follower count doesn't matter - Evidence: "You never know who's actually looking at that and who's reading" #### โฑ๏ธ [30:15](https://youtu.be/I9zQos_ePDs?t=1815) โ€” AI as learning tool, not replacement - Key idea: Use AI for conversation and skill gaps, not resume generation - Evidence: "Use them more as leveraging for you to kind of fill up the gaps" #### โฑ๏ธ [35:00](https://youtu.be/I9zQos_ePDs?t=2100) โ€” Positioning AI usage positively - Key idea: Frame AI experience as architectural decision-making enhancement - Evidence: "I used it leveraging my architecture decisions... cut the production time" #### โฑ๏ธ [38:40](https://youtu.be/I9zQos_ePDs?t=2320) โ€” GitHub features for career growth - Key idea: Profile README, GitHub Actions, and Discussions board optimization - Evidence: "I look at the GitHub actions workflow tabs in your repositories" #### โฑ๏ธ [42:20](https://youtu.be/I9zQos_ePDs?t=2540) โ€” Technology trends for 2026 - Key idea: Focus on infrastructure as code and systems programming - Evidence: "Growth of infrastructure as code practices... that's the way the world is spinning" ## ๐Ÿ“š References - [GitHub State of the Octoverse Report](https://github.com) - [GitHub Discussions Community](https://github.com/discussions) - [Just Another Job Seeking App (JAJA) - Open Source Project](https://github.com) - [Anthropic Career Guidelines](https://anthropic.com/careers) ## ๐Ÿ—ฃ๏ธ Paraphrase - Junior developer roles have evolved rather than disappeared - AI has raised the entry bar by making basic coding accessible to everyone, so new developers need broader systems thinking including debugging, security, and architecture skills from the start - Strategic GitHub profile optimization matters more than activity volume - recruiters scan profiles in 10 seconds focusing on pinned repositories, bio clarity, and meaningful contribution patterns rather than green square quantities - Effective AI usage involves leveraging it as a learning conversation partner and architectural decision aid while maintaining fundamental skills and framing AI experience positively on resumes rather than hiding or over-depending on it ## ๐Ÿ“ฐ Newsletter Ready to level up your developer career in 2026? ๐Ÿš€ GitHub Developer Advocate Andrea Griffiths shares game-changing insights about the evolving tech job market. The key takeaway? Junior developer roles aren't disappearing - they're transforming. AI has raised the bar, making basic coding skills table stakes while demanding deeper understanding of systems, security, and architecture. **Your career action plan:** - Optimize your GitHub profile strategically (pinned repos > green squares) - Use AI as a learning partner, not a replacement for fundamentals - Build authentic community presence through public documentation - Focus on infrastructure as code and systems programming trends The WordPress community can especially benefit from these insights as we navigate our own ecosystem changes. Consider how these strategies apply to contributing to Core, building plugins, or advancing your WordPress development career. Resources: [GitHub Discussions](https://github.com/discussions), [State of the Octoverse](https://github.blog), [Learn WordPress](https://learn.wordpress.org/), [Developer Docs](https://developer.wordpress.org/) ## โœ… Action Items #### DevRel - Developer advocacy: Create GitHub profile optimization workshop series with live examples - Education: Host office hours on AI usage best practices for career development - Developer experience: Document GitHub features for career advancement (Actions, Discussions) - Community management: Moderate discussions around career advice and job market trends #### WordPress - Core: Encourage meaningful GitHub activity patterns for contributors - Community: Create mentorship programs pairing experienced developers with career changers - Training: Develop lesson plans on systems thinking and architecture for junior developers - Documentation: Style guides for professional GitHub presence and portfolio development #### Community - Maintainers: Support authentic contribution recognition beyond commit counts - Contributors: Onboarding programs emphasizing public learning and documentation - Sponsors: Fund career development initiatives for underrepresented developers ## ๐Ÿงช Hypotheses - Developers who document their learning journey publicly have higher hiring success rates than those who code privately - GitHub profiles with strategically curated pinned repositories receive more recruiter attention than profiles with numerous unorganized repos - Career changers who leverage domain expertise alongside technical skills outperform traditional CS graduates in certain roles - Companies increasingly value AI-assisted development skills when presented as architectural decision-making tools ### ๐Ÿ“Š Methodology - Track hiring outcomes for developers who maintain public learning documentation vs. private portfolios - A/B test GitHub profiles with strategic curation vs. comprehensive showcasing - Survey recruiters on profile scanning patterns and decision factors - Analyze correlation between community participation and career advancement ### โ“ Questions - How do hiring managers differentiate between authentic AI usage and over-reliance during technical interviews? - What specific community contributions create the strongest networking effects for career advancement? - How should career changers balance domain expertise with technical skill development in competitive markets? ### ๐Ÿ”ฎ Future Research - Impact of AI coding assistants on junior developer skill development pathways - Effectiveness of different professional networking strategies in remote-first tech culture - Evolution of technical interview processes as AI tools become standard development practice ### ๐Ÿ’ก Implications - **Practice**: Shift from quantity-focused coding practice to quality-focused systems thinking and documentation - **Education**: Update bootcamp and university curricula to emphasize architecture and debugging alongside basic programming - **Hiring**: Recruiter training on meaningful GitHub activity patterns vs. superficial metrics - **Career Development**: Investment in public learning platforms and community participation tools ### ๐Ÿง  Learning Styles Assessment Reading level estimate: 10th-11th grade (accessible to most developers) #### ๐Ÿ‘€ Visual learners - Present: Speaker video, occasional screen sharing of GitHub profiles - Improve: Diagrams of career pathway progression, visual GitHub optimization examples #### ๐Ÿ”Š Aural learners - Present: Full audio conversation with clear dialogue - Improve: Timestamped audio summaries, speaker notes for key concepts #### ๐Ÿ—ฃ๏ธ Verbal learners - Present: Clear definitions of terms like "infrastructure as code", practical examples - Improve: Career terminology glossary, analogies for technical concepts #### ๐Ÿงฉ Kinesthetic learners - Present: Actionable GitHub profile optimization steps - Improve: Interactive GitHub audit checklists, hands-on portfolio exercises #### ๐Ÿง  Logical learners - Present: Structured argument progression, cause-effect relationships - Improve: Career decision flowcharts, hiring process timelines #### ๐Ÿง‘โ€๐Ÿคโ€๐Ÿง‘ Social learners - Present: Community-focused advice, networking emphasis - Improve: Discussion prompts for developer communities, peer review guidelines #### ๐Ÿง˜โ€โ™€๏ธ Solitary learners - Present: Self-reflection opportunities on career direction - Improve: Personal career assessment worksheets, individual skill development paths ### ๐Ÿ“Š Accessibility Assessment #### Screen reader semantics - Present: Clear speaker transitions, descriptive context - Improve: Transcript with speaker labels, heading structure for easy navigation #### Keyboard navigation - Present: Standard YouTube player controls - Improve: Timestamp links for easy topic jumping #### Contrast and readability - Present: Good video quality with clear speakers - Improve: High contrast transcript option, adjustable text sizing #### Media and captions - Present: Auto-generated captions available - Improve: Professional captions with technical term accuracy, downloadable transcript ### ๐ŸŒ DEIB Assessment #### Cultural breadth - Present: Career changer perspective, non-traditional pathway discussion - Improve: Multiple geographic hiring market examples, international developer experiences #### Inclusive language - Present: Gender-neutral role descriptions, welcoming tone for beginners - Improve: Avoid industry jargon without definition, provide context for technical terms #### Representation - Present: Female developer advocate sharing experiences - Improve: Acknowledge different learning styles, career change motivations #### Psychological safety - Present: Supportive tone about job market challenges, realistic expectations - Improve: Content warnings about layoff discussions, resource links for support ### โœ… Logical Validity #### Strong reasoning - Deductive strength: "AI makes basic coding accessible โ†’ higher entry barriers โ†’ need for systems thinking" [09:15] - Inductive strength: Personal experience and recruiting observations support profile optimization advice [15:20] - Causal reasoning: Clear connection between community participation and hiring opportunities [18:45] #### Fallacies - Appeal to authority: Minimal - relies on experience rather than credentials - Hasty generalization: Some broad claims about "everyone uses AI" lack nuance [30:15] #### Tone of voice - Overall tone: Encouraging and realistic, balances optimism with practical challenges ### ๐Ÿ›ก๏ธ DARVO Analysis - Deny: No defensive patterns observed - Attack: No attacks on accusers or critics - Reverse victim/offender: No victim/offender role reversals - Emotional cues: Genuine empathy for job seekers, authentic career sharing ## ๐ŸŽฏ Conclusion The developer career landscape has fundamentally shifted - junior roles require more comprehensive skills while AI becomes both challenge and opportunity. Success in 2026 depends on strategic GitHub presence, authentic community engagement, and positioning AI as an architectural enhancement tool rather than a replacement for fundamental skills. Focus on systems thinking, public learning documentation, and building genuine professional relationships through community participation. ## ๐Ÿงฑ Refactor Appendix ### ๐Ÿ“š Literature Split Pad Split level: H3 ### [H1] Junior developer roles require systems thinking beyond basic coding > [!quote] Dev Career Secrets from a GitHub Strategy Leader by Andrea Griffiths via YouTube. Published 2024-12-28. > "Can you debug that app? Can you put proper security measures in place?" AI accessibility has raised entry-level expectations requiring architecture knowledge. [view highlight](https://youtu.be/I9zQos_ePDs?t=720) #career #junior-dev ID:rw_001 ### [H2] GitHub contribution graphs mislead hiring decisions > [!quote] Dev Career Secrets from a GitHub Strategy Leader > "I don't judge a person's capability by how many green squares" Quality of contributions matters more than commit frequency for evaluation. [view highlight](https://youtu.be/I9zQos_ePDs?t=920) #hiring #github ID:rw_002 --- ### ๐Ÿ”– Atomic Split Pad (Zettel Candidates) Split level: H4 #### [Z1] Public learning documentation creates career opportunities - Applies to developers building professional networks - Referenced at timestamp 22:10 - Links: [[Community-Building]] [[Career-Development]] #### [Z2] AI usage requires strategic positioning in job applications - Relevant for developers incorporating AI tools in workflows - Referenced at timestamp 35:00 - Links: [[AI-Strategy]] [[Resume-Optimization]] --- ## Full Transcript [00:00] the talk with Franchesco series and I<br>[00:03] have of course a super guest. Not not a<br>[00:06] special guest but a super special guest.<br>[00:09] Usually when I have people here on the<br>[00:11] podcast first we do the podcast and then<br>[00:14] we we meet in person and then we take a<br>[00:16] selfie with with Andrea. We did<br>[00:18] something. So one second. Okay. And um<br>[00:22] in this case we did the opposite. So we<br>[00:25] met in person multiple times. I think<br>[00:27] that Andrea almost got um let's say<br>[00:30] bored of me last year. But [laughter] we<br>[00:32] are here [gasps] today of course to talk<br>[00:35] about developer careers 2026. I can't<br>[00:38] believe that we are at the end of 20125<br>[00:40] that in three two weeks and half will be<br>[00:43] straight into 2026. That's mind-blowing.<br>[00:46] Welcome Andrea. How are you? Please<br>[00:48] introduce yourself.<br>[00:50] >> Hello Francesco. I could never be bored<br>[00:52] of you. You're my favorite Italian. So<br>[00:55] that would be a lie. No. Hi. Hi. I'm<br>[00:58] Andrea Griffiths. I am a developer<br>[01:01] advocate at GitHub and I did have the<br>[01:04] pleasure of seeing Franchesco<br>[01:07] three times last year. Was it three<br>[01:09] times? Yeah.<br>[01:10] >> Uh yeah. So it was great to meet you. I<br>[01:13] mean obviously like I've I knew of you.<br>[01:15] I followed you online like a lot of us.<br>[01:18] Uh but it was so nice every time y'all<br>[01:22] if you're watching this and you know<br>[01:23] what I'm talking about. Every time I<br>[01:24] meet someone in person that I known<br>[01:26] online, it's always kind of a little bit<br>[01:29] scary because I have been disappointed.<br>[01:32] They say don't meet your heroes and some<br>[01:34] people are just not who they show they<br>[01:37] are online. But you are the real deal.<br>[01:40] So I've been veryful.<br>[01:42] >> Of course she's forced to do to to say<br>[01:44] this now because she is on my podcast.<br>[01:47] But I will keep an eye on this if she<br>[01:49] calls me out. You know I don't you<br>[01:51] [laughter] know I don't I can't I can't<br>[01:52] lie to our amazing friends here. No<br>[01:55] chance.<br>[01:55] >> Perfect. Perfect. Before we keep going.<br>[01:58] Hi Razan. Hey Yorkdev. How are you? How<br>[02:00] is it going? Yes of course we have. By<br>[02:03] the way it's by chance we are wearing<br>[02:05] zers by chance but hair is way way<br>[02:08] better than me. Of course. So Andrea uh<br>[02:12] a little bit about yourself and then I<br>[02:14] want to really focus this episode for us<br>[02:17] for us developers because I think it has<br>[02:19] been a tough year for most of us<br>[02:21] layoffs. So I want to try to give say<br>[02:23] something that can help as many people<br>[02:26] as possible. Feel free to drop questions<br>[02:29] related to career um getting a job in<br>[02:32] tech. I have a list of questions. If we<br>[02:34] have some very cool questions we can<br>[02:36] also pick them. Okay, Andrea, please<br>[02:40] tell us a bit more about yourself.<br>[02:42] >> Yeah, so I I get to work at GitHub where<br>[02:45] it's kind of a wild wild thing to even<br>[02:49] think that I I get to work there and I<br>[02:50] get to support all of you. Uh which is<br>[02:53] great. I'm I'm I'm in a role that I<br>[02:54] really enjoy. Uh but I didn't start my<br>[02:58] path to uh computer engineering in a<br>[03:03] traditional way. I'm a career changer.<br>[03:05] her church careers later in life. So,<br>[03:08] it's an interesting it's been an<br>[03:10] interesting process. Um, but I would say<br>[03:13] now I'm not so new new, you know, like<br>[03:15] I've been doing this now for almost 10<br>[03:17] years. Uh, but it's definitely<br>[03:19] interesting to get started. Um,<br>[03:21] especially now, I think if I was trying<br>[03:23] to make that career shift now, it would<br>[03:26] probably be a lot harder than it was for<br>[03:28] me back in 2016 because times are very<br>[03:32] different. like new developers, even a<br>[03:35] senior developers are facing a<br>[03:37] completely new market. We're on the what<br>[03:39] is it the fourth industrial revolution<br>[03:42] at this point and times are moving fast.<br>[03:45] A lot of companies are making a lot of<br>[03:47] changes. A lot of companies love to do<br>[03:48] this right before the holidays. So,<br>[03:50] shout out to all of you out there. If<br>[03:52] you're being affected by recent layoffs,<br>[03:54] I'm so sorry. Like, please connect with<br>[03:56] me via LinkedIn or here. And if there is<br>[03:59] any opportunity that I see out there<br>[04:00] that I can share with you, I'll be happy<br>[04:02] to. So there's a lot of uncertainty in<br>[04:04] our market. One thing that I think is<br>[04:06] not uncertain is that we need developers<br>[04:09] more than ever. So hopefully the<br>[04:11] conversation that we have today kind of<br>[04:13] helps people get a little less doom and<br>[04:17] gloomy because I think we're all kind of<br>[04:19] a bit exhausted of AI, exhausted of<br>[04:23] everything that's going to hell, of all<br>[04:25] the negativity. So hopefully we can walk<br>[04:28] out of here today feeling like, okay,<br>[04:30] I'm here. There's a place for me in this<br>[04:33] industry and I can be successful. I have<br>[04:35] a very non-traditional background and<br>[04:37] I've been able to grow a career that I'm<br>[04:40] proud of. Keeping it authentic, not<br>[04:42] telling lies about who I like on the<br>[04:45] internet. That's for sure, Franchesco. I<br>[04:47] don't lie. My face won't allow me to.<br>[04:49] You'll be able to see right through me.<br>[04:50] [laughter]<br>[04:51] So, yeah, that's a little bit about<br>[04:53] myself.<br>[04:55] >> Perfect. Thank you so much. And yes, I<br>[04:56] think we are all all a bit exhausted of<br>[05:00] so much AI. Not because AI is so bad,<br>[05:02] but I remember that I for some reasons I<br>[05:05] brought a talk I think last year that<br>[05:07] was not about AI because I think it was<br>[05:09] Rust and web development. So I I started<br>[05:12] saying like I'm sorry but my talk is not<br>[05:14] about AI and everybody laughed in the<br>[05:16] audience of course. So this made me feel<br>[05:19] in person like what is the real feeling<br>[05:23] of developers. I think a great plus of<br>[05:25] going to conference is that you know<br>[05:26] it's not Twitter so you have you know<br>[05:28] the real vibes the real feeling of what<br>[05:30] people feel but okay so<br>[05:32] >> for sure<br>[05:33] >> yes Andrea let's start with addressing<br>[05:37] you know the elephant in the room I want<br>[05:39] to start with the super spicy question<br>[05:42] which is like<br>[05:44] is the role of junior developer like uh<br>[05:46] like doesn't exist anymore like what's<br>[05:49] happening with the junior developers and<br>[05:52] this is I think the most important thing<br>[05:55] that I want to discuss with you because<br>[05:58] uh I think the barrier now is higher. I<br>[06:00] would like to know from you what what do<br>[06:03] you think about this you know the role<br>[06:06] of the junior developer? Yeah. No, no<br>[06:08] way. If anything, there is there is a<br>[06:11] need for more developers for sure. And<br>[06:14] hi friends. Um, contrary to what some<br>[06:17] people on the internet, especially on<br>[06:19] Twitter, will have you believe, no one<br>[06:21] is born a senior engineer. [laughter]<br>[06:23] Like every single person had to start at<br>[06:25] some point as a junior. So, I don't<br>[06:28] think the role is disappearing as much<br>[06:30] as it is going through a shape-shifting<br>[06:35] moment, right? Like it's changing.<br>[06:37] uh the way that we're<br>[06:40] creating software now has changed. So<br>[06:43] that's obviously going to change that<br>[06:44] entry point for a lot of people, right?<br>[06:46] So like you remember your first scenes<br>[06:49] that you used to do when you the first<br>[06:51] from like hello world, right? Like the<br>[06:53] first lines of code where you were like,<br>[06:54] "Wow, I I made a machine do something.<br>[06:56] Look at that." Uh and it's like great,<br>[06:59] you you can write for a loop, right? But<br>[07:02] now there is a higher bar. Um AI,<br>[07:05] believe it or not, has raised the way<br>[07:07] that we have to think about what it<br>[07:09] means to be a junior developer. And so<br>[07:12] that old part of like, okay, yeah,<br>[07:14] you're able to write code and you can<br>[07:18] sit down and hack it out and here's a<br>[07:21] page like it's a bit irrelevant now<br>[07:23] because there is a lot of systems in<br>[07:24] place that made that entry point sort of<br>[07:27] table stakes. So now we can everybody<br>[07:31] can go and b code an app basically<br>[07:33] pretty much we're using the the right<br>[07:35] tools. Uh the question is can you debug<br>[07:38] that app? Can you put proper security<br>[07:42] measures in place in that app so that<br>[07:44] you're not losing your entire database<br>[07:46] of customers if you're, you know, like<br>[07:48] we've seen the horror stories play out<br>[07:49] on tech Twitter, on Reddit of people<br>[07:52] that vipcoded things and then they're<br>[07:54] like, "Oh no." Uh, or even catastrophic<br>[07:57] where it's like you lose information,<br>[07:58] you're supposed people's information or<br>[08:01] you're getting a $50,000 AWS bill<br>[08:05] because you post a key somewhere. So I<br>[08:07] think AI sort of raised the bar in what<br>[08:10] it means to be a junior developer and<br>[08:12] that we're forced now to go into the<br>[08:13] field with a lot more context and<br>[08:16] thinking a lot more about okay this is<br>[08:19] not just the low entry point table<br>[08:22] stakes like I have to actually know why<br>[08:24] things work. I have to actually know how<br>[08:27] to debug them. And I have to think more<br>[08:29] as an architect and as a systems person<br>[08:31] like design systems are so important<br>[08:33] right now. Uh because again anyone can<br>[08:36] go in and hop in one of the many many<br>[08:38] many tools that are out there to like<br>[08:41] just create a skeleton app. But if you<br>[08:43] want to create something that truly<br>[08:45] scales and something that's going to<br>[08:47] support a business or you're going to go<br>[08:48] work for a business supporting their<br>[08:50] applications like you need to actually<br>[08:52] know what you're doing. So it's<br>[08:53] morphing. I think is shapeshifted. Um AI<br>[08:57] can do a lot of things for us. It can<br>[08:59] create these beautiful scaffoldings. It<br>[09:01] can autocomplete, but again it's not<br>[09:04] going to make those important decisions<br>[09:06] that really matter at the end like the<br>[09:08] security, the architecture of the things<br>[09:09] that you're building. So I think for<br>[09:12] junior developers right now more than<br>[09:13] ever it's important that we sort of take<br>[09:16] more ownership uh of of software<br>[09:18] development as a whole. Um, it used to<br>[09:21] be that when you came in as a junior<br>[09:23] developer in the company, you were very<br>[09:24] pigeon hole like okay, you're the<br>[09:26] front-end person, you're going to be<br>[09:27] focused on this part and now like you<br>[09:30] really need to have a broader look, a<br>[09:33] bigger understanding of everything. Um,<br>[09:36] a bigger understanding of what it means<br>[09:38] to the business really things that we<br>[09:40] often don't think about because we're<br>[09:41] like, no, my job is to hack it out and<br>[09:43] write code and that's what I do. But to<br>[09:45] be really successful now, you have to<br>[09:47] have sort of that eagle view and take<br>[09:49] all those things into consideration. So<br>[09:51] I don't think it's it's going to be a<br>[09:53] career that is expiring as much as it's<br>[09:56] a career that is changed a lot and now<br>[09:58] you have to come into the field with a<br>[10:00] lot more skill and thinking more broadly<br>[10:03] than before. um you know like even<br>[10:05] internships like interns are doing<br>[10:07] things now that before I I came to<br>[10:09] GitHub the first time as an apprentice<br>[10:10] and like I definitely did not have uh<br>[10:14] even the opportunity to be as impactful<br>[10:16] as some of like our summer interns are<br>[10:17] doing things that are like going into<br>[10:19] production changing the business so it's<br>[10:21] it's it's a exciting time in a way I<br>[10:24] think for juniors because you can be a<br>[10:25] lot more impactful but it's also a<br>[10:27] harder time because you do have to come<br>[10:29] more prepared like you can't the time<br>[10:30] for learning the table stakes things is<br>[10:33] is it's got to before you enter the<br>[10:35] market.<br>[10:37] [snorts]<br>[10:37] >> Yes, thank you. Thank you, Andrea. And I<br>[10:39] think that this visualizes so all the<br>[10:43] say most of the problems that are<br>[10:45] actually in the in the tech ecosystem.<br>[10:49] So AI made basically everything easier<br>[10:51] and everything harder. So we are at the<br>[10:54] starting point again because this is how<br>[10:56] it works because it's an ecosystem. And<br>[10:58] so if everything if everybody can bible<br>[11:00] code an app then probably a person who<br>[11:03] knows how to code properly will still<br>[11:04] make a better app than you. We have<br>[11:07] sharper knives but the chefs still cooks<br>[11:09] better than everyone else because for<br>[11:11] them this is a it's a job. Absolutely.<br>[11:15] Yes. So I think we I think we found a<br>[11:19] solution Andrea which is we just use uh<br>[11:22] AI to have a whole green GitHub profile<br>[11:27] and then we are done. We'll we just get<br>[11:29] hired right what do you think about this<br>[11:32] solution? No, Franchesco. No. And<br>[11:34] honestly, and I I say this from the<br>[11:37] bottom of my heart, like I I don't take<br>[11:39] that into consideration, and I I've I've<br>[11:42] been very lucky that I have some I I've<br>[11:45] been I've been able I have friendships<br>[11:47] that I cultivated throughout the years<br>[11:48] with both recruiters that are recruiters<br>[11:50] who are non-technical at all and they're<br>[11:53] just looking at like, okay, yeah, when<br>[11:55] they get thrown a profile of someone,<br>[11:57] they look at certain things. And then<br>[11:58] some recruiters who actually have a<br>[12:00] technical background and and would know<br>[12:01] what they're looking at when they look<br>[12:02] at something like your contribution<br>[12:04] graph, right? And honestly, that is<br>[12:06] never told the entire story. I don't<br>[12:09] judge a person's capability by how many<br>[12:11] green squares. I mean, as a matter of<br>[12:14] fact, I might even judge you a little,<br>[12:15] but not in a good way. Like, if you show<br>[12:17] me like something that is [laughter]<br>[12:20] super like panon green all the way, be<br>[12:22] like, "Wow, when where is where is the<br>[12:25] rest of you?" Like,<br>[12:28] if this is all there is like where is<br>[12:30] there more to the person? So I don't<br>[12:32] think it's never told a full story. It's<br>[12:34] told a story but it's still a really<br>[12:36] unreliable story because people gamify<br>[12:39] it too like you know they they commit I<br>[12:41] mean yeah and anyone can do something to<br>[12:43] commit so there is a a a square in your<br>[12:45] contribution graph. um I look at more<br>[12:49] like real signals like the meaningful<br>[12:52] contributions and the pattern<br>[12:54] contribution is not always 100% code and<br>[12:57] that's another thing that I think it's<br>[12:58] also very misunderstood like yes it's<br>[13:01] great that I can see this brilliant<br>[13:03] project that you created and all this<br>[13:06] obviously you were busy the month of<br>[13:08] December because that's when all your<br>[13:09] green squares are happening um but then<br>[13:12] maybe you open source it maybe you<br>[13:14] didn't maybe you work in a team maybe<br>[13:15] you didn't I mean looking at things like<br>[13:17] issue discussions like the way that<br>[13:19] you're actually communicating the value<br>[13:20] of your project, how it's documented,<br>[13:23] what it is read of your project looks<br>[13:24] like, uh how your code like your<br>[13:27] commits, what what is that conversation<br>[13:30] happening even when it's a solo project<br>[13:32] like how are you yourself doing<br>[13:35] engineering best practices when you're<br>[13:37] approaching a project even if it's just<br>[13:38] your own project. So I don't think it<br>[13:40] takes the whole story. I think I hate it<br>[13:42] that I love it because it's such a cool<br>[13:44] thing. And listen, I'm a fan of the<br>[13:45] contribution grab. I got the mug<br>[13:48] >> to show it to prove to prove it. I want<br>[13:50] it. But but I don't I don't think it<br>[13:52] tells the full story though. Honestly, I<br>[13:54] don't think I I don't think the gold<br>[13:55] standard is is is<br>[13:58] >> quantity is quality should be value. Um<br>[14:01] it's the same thing like if you see<br>[14:03] people that are online posting a ton of<br>[14:05] content, making a lot of tweets, and<br>[14:08] what are they really saying? they're<br>[14:10] saying nothing. Um, so I I I don't judge<br>[14:13] that as a good um bar to say this is a<br>[14:17] better engineer than the other. Um, I<br>[14:21] think looking through the way that<br>[14:23] you're reviewing a PR gives me a gold<br>[14:25] mine of information. Are you giving<br>[14:27] thoughtful feedback? Is the interactions<br>[14:30] where like and again the way that you're<br>[14:31] documenting this is so important because<br>[14:33] it lets me know like if you're able to<br>[14:34] explain the way that you're designing<br>[14:36] something, you'll be able to explain it<br>[14:37] to the team. you'll be able to work with<br>[14:39] the team. If you if you're a person<br>[14:41] that's actually building in public and<br>[14:43] you're contributing to open source<br>[14:44] projects, like you better believe it.<br>[14:46] I'm going to go back and dig through<br>[14:47] those and take a look and see how those<br>[14:50] interactions happen. Um especially when<br>[14:52] I'm I'm in a position where I can give<br>[14:54] an opinion of what I hire because that<br>[14:56] to me says more, you know, are you a<br>[14:57] part of a developer community? Like how<br>[15:00] how are you interacting there? How are<br>[15:02] you treating people there? Um so even as<br>[15:06] like PRs and your contributions the<br>[15:09] merge PRs like still there is so many<br>[15:11] things that they don't fully show. Um<br>[15:15] but I think that there there is a room<br>[15:17] for us to look at the contribution graph<br>[15:19] as a part of a person but not the whole<br>[15:21] story. So um I definitely don't don't<br>[15:26] judge the quality of of someone's<br>[15:28] contributions just because it looks like<br>[15:30] they made the most. Uh it's uh it's a<br>[15:33] sad thing that you know it's the same<br>[15:34] story that and please don't don't add me<br>[15:37] that people that are like oh and they<br>[15:39] love to post pictures of like it's 11<br>[15:41] o'clock at night and everybody's still<br>[15:42] in the office. I'm like holy crap that<br>[15:44] tells me I never want to work for you.<br>[15:46] It doesn't mean I didn't work at 11<br>[15:48] o'clock sometimes or on the weekends or<br>[15:51] you know like I do because I do I I do<br>[15:53] take a lot of pride in my work and I do<br>[15:54] put myself in positions where it's like<br>[15:56] ah why'd you do it? But but um that that<br>[15:59] shouldn't be something that's glorified,<br>[16:01] you know. Um and it doesn't tell the<br>[16:03] full picture of like the quality of<br>[16:04] person you are both in engineering and<br>[16:06] in life. So<br>[16:08] >> yeah,<br>[16:09] >> exactly. I'm definitely sleeping at 11<br>[16:11] p.m. But then of course I still can do a<br>[16:13] lot of stuff. I can wake up early, but<br>[16:15] it's not something that should be<br>[16:17] glorified. I'm always a huge s sustainer<br>[16:20] of sustainability<br>[16:22] and um because this is what will let you<br>[16:24] stay in if you want to stay in this<br>[16:26] career for 30 years, 40 years depending<br>[16:28] on your age but I want to stay here for<br>[16:31] even after retirement. So<br>[16:34] perfect. Thank you. Thank you so much.<br>[16:36] Thank you so much Andrea. Now let's make<br>[16:38] a quick h quick game. You are a<br>[16:42] recruiter. You have 10 seconds to check<br>[16:45] one profile because you have to check a<br>[16:47] thousand profiles at 100 profile and you<br>[16:50] have to dedicate just 10 seconds to each<br>[16:53] of them because you need you know you<br>[16:55] need to filter a bit because you want to<br>[16:56] have a life right. So what would you do?<br>[17:00] I think this is this information can<br>[17:03] help you know to at least be sure that<br>[17:06] you are at least checking this. So what<br>[17:08] is the thing you will do you know to say<br>[17:10] you know the yes no you know the X<br>[17:12] factor thing in 10 seconds. Okay. So, I<br>[17:16] land on your GitHub profile right now<br>[17:18] and uh first thing I look at uh not the<br>[17:22] contribution graph. Um your pin repos,<br>[17:25] right? Like the repositories that you<br>[17:27] pin. Uh<br>[17:30] what what are you choosing to show me?<br>[17:32] That's that tells me a lot about what's<br>[17:34] important to you, right? Whether it be<br>[17:35] your your own projects or projects that<br>[17:37] you forked and you're working on. I<br>[17:39] mean, I think that that says a lot about<br>[17:40] what your priorities are. Um your bio,<br>[17:44] right? Uh even that short little<br>[17:46] sentence on the about me kind of tells<br>[17:48] me what sort of how you want to present<br>[17:51] yourself and that's going to that's<br>[17:52] going to tell me a lot about who you are<br>[17:54] as a developer. And then I do look at<br>[17:57] the recent activity but then again not<br>[18:00] the volume of the recent activity but<br>[18:02] the type of the recent activity. So are<br>[18:04] you opening issues? Are you reviewing<br>[18:06] PRs or are you just pushing commits to<br>[18:09] get these green squares to repose that<br>[18:12] nobody's actually going to see? Um I I<br>[18:16] would look at those things first and<br>[18:19] then that's like sort of looking for the<br>[18:21] the the entry point positive things but<br>[18:23] then there are some like okay like I<br>[18:26] will move immediately I guess from the<br>[18:28] from the profile if there is nothing if<br>[18:30] there is no information. And I know a<br>[18:32] lot of people are very there's still if<br>[18:34] you're coming into the industry, you're<br>[18:35] new, you're just setting up your GitHub<br>[18:36] profile, like there is literally you can<br>[18:38] take 10 minutes of your time to write an<br>[18:41] about you, write your readme profile um<br>[18:44] and just give a little bit more context<br>[18:45] on who you are because whether we like<br>[18:46] it or not, recruiters do look at this,<br>[18:48] right? So, uh, the readmi profiles,<br>[18:52] there's a lot of really amazing ones out<br>[18:54] you can use for reference, but also<br>[18:57] there is some brand of like really icky<br>[18:59] kind of templated clearly copy and<br>[19:01] pasted and not giving attribution to the<br>[19:03] people who actually created them. So,<br>[19:05] that would be a turnoff for me. I'll be<br>[19:06] like, h never mind. Uh, and then a ton<br>[19:09] of like fork repos that just show like,<br>[19:12] okay, you a hundred different things,<br>[19:14] but nothing really stuck. like there<br>[19:17] were there was nothing that actually<br>[19:18] showed up on the recent activity. Um and<br>[19:21] then because I'm me and I again back to<br>[19:24] the I want you to come as a whole person<br>[19:27] like I would love to know a little bit<br>[19:28] more about you beyond the code. Um, so<br>[19:32] maybe in if your private life is<br>[19:34] private, but you know, like what you're<br>[19:36] passionate about coding and and yeah,<br>[19:38] you're super passionate about type<br>[19:40] languages. Like I love that, but also<br>[19:42] like throw me a bone. Tell me you like<br>[19:43] to ride a bike on the weekend or<br>[19:45] something like those kind of things kind<br>[19:47] of I I find very interesting. So, um, an<br>[19:50] empty profile I probably forgive quicker<br>[19:52] than a busy for the sake of busy sake<br>[19:55] one, but take a time to like make a<br>[19:58] profile that, you know, at least tells<br>[19:59] me something about you at first glance.<br>[20:02] >> Perfect. And of course, does it mean<br>[20:05] that if you have a super polished GitHub<br>[20:07] profile that you are the best developer<br>[20:09] in the world?<br>[20:10] >> No. But you are you are present you this<br>[20:12] is how you present yourself in so in<br>[20:14] these difficult times. If there were you<br>[20:16] know job for everyone, everyone had know<br>[20:19] time to make long interviews but this is<br>[20:21] not how it works and this is why it's so<br>[20:25] important. Think if you are here in the<br>[20:27] chat and you have coded a lot but you<br>[20:30] never p pushed on GitHub you are in<br>[20:33] trouble because it's the same thing of<br>[20:36] um playing guitar 10 10 hours a day but<br>[20:39] you never do anything. How are you<br>[20:41] people supposed to know that you are<br>[20:44] that good? That's the real That's the<br>[20:46] real issue. So, thanks. And I liked<br>[20:49] especially the the pinned part that I<br>[20:51] think is absolutely underestimated. I<br>[20:53] think that if you're looking for a React<br>[20:54] role, you should have at least two<br>[20:57] GitHub repositories pinned that are<br>[20:59] about React because this is what<br>[21:01] >> exactly<br>[21:02] >> you should work because exactly this is<br>[21:04] where people check. It's the same reason<br>[21:05] why people check the pin tweet or what<br>[21:08] you post on YouTube on the first the<br>[21:10] first video. So the people check the<br>[21:12] pins and stuff because they don't have<br>[21:14] time maybe to check 100 repositories<br>[21:17] and uh yes and also and also I liked the<br>[21:20] fact not too many things if you are an<br>[21:23] expert in everything you are an expert<br>[21:24] in nothing I remember my first CV was<br>[21:27] that long no probably I mentioned two<br>[21:29] things last JavaScript that's it because<br>[21:33] now yes exactly because now now you know<br>[21:36] exactly what you want to work on<br>[21:39] >> you can say I can do everything you be<br>[21:42] you your value will be perceived as as a<br>[21:46] as a lesser. Maybe you are an expert<br>[21:48] better than everyone else on on many<br>[21:50] technologies but still maybe you are<br>[21:52] looking for something specific not just<br>[21:55] please give me any random job in tech.<br>[21:58] This is not how it works. Usually<br>[21:59] companies they look for a specific role<br>[22:02] react developer backend developer with a<br>[22:05] specific stack a specific library. This<br>[22:07] is why they'll give you money. This is<br>[22:08] why they give you a salary. So this is<br>[22:11] uh this is something like that I think<br>[22:13] it's important to mention. Thank you.<br>[22:16] >> Thank you. So maybe we can't uh we're<br>[22:20] never sure if we will get hired because<br>[22:21] this can also depend on the company they<br>[22:25] needs maybe they listed they need a<br>[22:27] Python and JavaScript developers but in<br>[22:29] reality they need a Python developer. I<br>[22:30] don't I have no idea why this happens<br>[22:32] but this is the truth. So how can we<br>[22:35] make ourselves hierallable? So it's like<br>[22:38] okay like we can get we can get hired.<br>[22:41] So basically to increase our chances.<br>[22:46] >> That is excellent an excellent excellent<br>[22:48] question. Um [clears throat] I mean I<br>[22:51] think it that really sort of depends of<br>[22:53] what stage of your career you're in<br>[22:55] right like where where you are. But if<br>[22:58] we're talking about for folks that are<br>[23:00] just new to the industry like they're<br>[23:01] trying to come in as junior devs. Um<br>[23:05] obviously the things that we mentioned<br>[23:07] earlier like having that good profile<br>[23:10] having well documented yes if you're<br>[23:12] applying for a job at Facebook to work<br>[23:15] on react I will hope that you either<br>[23:17] contributed or at least have one project<br>[23:20] that shows you you know like the skill<br>[23:22] is there. Um but also I think the the<br>[23:27] the networking and the community aspect<br>[23:30] of it is sometimes sort of under under<br>[23:33] hype. Uh, I think it's it's<br>[23:36] underestimated how if you're in a more<br>[23:38] senior position in your career, nine<br>[23:40] times out of 10, you're not even like<br>[23:42] really applying for jobs. They're just<br>[23:44] you're getting sicked out like because<br>[23:47] a lot of these things come through<br>[23:48] networking, you know, like somebody like<br>[23:50] I know I know a really good Russ guy<br>[23:52] now. So if I ever have a Russ question<br>[23:55] or I have or I see a Rust related role,<br>[23:58] I know exactly who I'm going to call.<br>[24:00] Um, so just kind of sort of picking the<br>[24:03] thing that you're known for and make<br>[24:06] sure that you're talking about it out<br>[24:07] there. And one thing I think uh we also<br>[24:11] underestimates ourselves is how many<br>[24:13] people are actually like reading and<br>[24:14] looking and watching the things that<br>[24:15] you're doing. Um, if you have a account<br>[24:19] on Twitter, forever Twitter, and you<br>[24:21] have 50 followers and you're passionate<br>[24:23] about talking, I don't know, whatever it<br>[24:25] is, like a remix, you think it's so<br>[24:27] cool. I want to talk about it all the<br>[24:29] time. go there and talk about it. You<br>[24:31] never know who's actually looking at<br>[24:32] that and who's reading and where that<br>[24:34] opportunity might come from. Uh oh, but<br>[24:36] I don't have a 100,000 followers. It<br>[24:38] doesn't matter. Honestly, I've been so<br>[24:40] surprised at some some of the comments<br>[24:42] that I got from both people that I met<br>[24:43] in person and then online where they're<br>[24:45] like they'll mention something very<br>[24:47] specific that I did many years ago even<br>[24:50] and I'm like oh my like it blows my<br>[24:52] mind. I'm like how in the world did that<br>[24:54] even reach you? Um, so I would, you<br>[24:56] know, figure out the thing that you want<br>[24:57] to focus on and start putting yourself<br>[24:59] out there as a person who who is clearly<br>[25:02] an expert in that field or is clearly<br>[25:04] learning in order to become an expert in<br>[25:06] that field. That's also another<br>[25:08] underestimated I think skill opportunity<br>[25:11] for you to storytell. Um, humans we love<br>[25:15] we love to follow a story. That's that's<br>[25:17] just the way it is. Like and I I love<br>[25:19] I'm I'm on the positive following the<br>[25:20] story, hyping you up. Like I want to see<br>[25:22] you succeed. there be haters out there<br>[25:24] that want to see people like not<br>[25:26] succeed, but regardless, we are glued to<br>[25:29] a story. So, folks that do a really good<br>[25:31] job at documenting how they're learning,<br>[25:33] um, tend to be a lot more successful and<br>[25:34] and getting the exposure they need to<br>[25:36] actually at least get through this once<br>[25:38] you get through the scanners, get<br>[25:40] through that initial recruiter<br>[25:41] interview, then you have something to<br>[25:43] show. Uh, and being part of a community<br>[25:45] is a is a easy but also very rewarding<br>[25:49] way that you can sort of put your best<br>[25:51] foot forward there. So, um, I would say<br>[25:54] document everything that you're doing.<br>[25:55] I'm very, it's very hard for me to do<br>[25:57] that. That's something that I don't<br>[25:58] enjoy doing particularly, but I try to<br>[26:00] be more disciplined. And you have<br>[26:02] something to say, say it. I don't care<br>[26:04] if 10 people saw your tweet or a million<br>[26:07] people saw your tweet. Um, you know,<br>[26:10] have opinions that are they're your<br>[26:11] opinions on them. And you know,<br>[26:13] sometimes they might be contrarian to<br>[26:15] what the coolest person with the most<br>[26:18] followers is saying, but if you feel<br>[26:21] confident in your skill and in your<br>[26:24] opinion, say it anyway. Say it anyway.<br>[26:26] Um, there might be a company out there<br>[26:29] that's looking for that very specific<br>[26:31] take on that thing. So, don't don't<br>[26:33] don't follow the herd. Just own that own<br>[26:36] all of you. Um, that's one way that at<br>[26:38] least I think uh it would be it would be<br>[26:40] good for you. And then have your things<br>[26:42] online tidy like make sure you have your<br>[26:44] GitHub uh with a readme profile. I think<br>[26:47] Franchesco you need to do a how-to on<br>[26:49] that or or host maybe online that'll be<br>[26:52] a cool little like a display of good<br>[26:54] readies because some folks have done a<br>[26:56] really good job at exposing how great<br>[26:58] they are and you can just land on their<br>[27:00] GitHub page and you get like the full<br>[27:02] view. They're really really good at<br>[27:04] that. My read me is I would say mid.<br>[27:06] It's not the best. I haven't put a ton<br>[27:08] of effort in it, but I feel like I got<br>[27:10] enough there that is like the things<br>[27:11] that matter to me uh that you see when<br>[27:14] you land on my page. So, yeah.<br>[27:16] >> Yeah. And I mean, it's absolutely fine.<br>[27:19] We should not get obsessed with our<br>[27:21] profiles 247. It depends. And when when<br>[27:23] we'll have, for example, now I have more<br>[27:25] free time. I'll probably check out the<br>[27:27] read me if it's up to date. But at least<br>[27:29] once in a while, you should check it. I<br>[27:31] love your portfolio. I love your<br>[27:33] personal websites website.<br>[27:35] >> Oh, cool. I really love it. I like it's<br>[27:38] very simple, but it's very on it's<br>[27:41] really on point. I'm about to<br>[27:43] restructure part of my of my portfolio<br>[27:46] so I can get even inspired by it. I<br>[27:48] really like like it. We will leave the<br>[27:49] link below so you can check it<br>[27:51] >> and uh yes. Yes,<br>[27:53] >> absolutely.<br>[27:55] Yes. So now let's talk again about AI.<br>[28:00] AI and developers because I think<br>[28:02] developers they have this sort of like<br>[28:04] know love and hate with AI because they<br>[28:07] want to use AI to improve their career<br>[28:10] even to apply I mean you can even use<br>[28:12] apply to<br>[28:13] >> reason but then you don't know where to<br>[28:16] stop when you should you know when<br>[28:19] should the human actually show so what<br>[28:22] do you think is the best use of AI for<br>[28:26] developers who want to get hired so<br>[28:29] basically to to improve their career,<br>[28:31] especially in in this process. What's<br>[28:33] the best way to use it? Let's just uh I<br>[28:37] don't know, let JGBT do everything and<br>[28:39] you'll answer every email by clicking<br>[28:41] the AI button.<br>[28:43] >> No, [laughter]<br>[28:45] no, no. I think and and you you you've<br>[28:47] said it already like if you're using it<br>[28:49] as a tool for learning and growing, like<br>[28:52] amazing, for sure. Um I think there is a<br>[28:57] lot of misconceptions out there like<br>[28:58] it's not all bad and even actually most<br>[29:01] a few months ago I was taking a look I<br>[29:03] had um a fantastic open source project<br>[29:05] on a open source Friday stream. They're<br>[29:07] called just not just a J just another<br>[29:11] jobseeking app. It's J AJA for short.<br>[29:15] I'm going to find the link and share it<br>[29:16] with you Franchesco so you can share<br>[29:17] because it's actually a phenomenal<br>[29:19] project. Um I'm sorry. Let me mute this<br>[29:22] dog real quick.<br>[29:24] >> It's fine. It's fine.<br>[29:27] >> I hope you're enjoying this uh this live<br>[29:30] event with Andrea. Andreas is really<br>[29:32] amazing.<br>[29:33] >> It's like you you just This is how<br>[29:37] >> Okay, you're muted now. You're muted<br>[29:41] now. I can't hear you. You're muted.<br>[29:47] You're muted.<br>[29:51] Don't know why.<br>[29:52] >> So sorry. Give me two seconds.<br>[29:54] >> No problem. No problem at all. In the<br>[29:56] meanwhile, I can check the chat. So let<br>[29:59] me know if you are here. I have a<br>[30:01] question by was coder and conventional<br>[30:05] background medicine, economics boot<br>[30:07] camps. Amazing. You have background in<br>[30:08] medicine. No degree in computer science.<br>[30:10] You can absolutely find I think a um job<br>[30:14] in um in computer in um as developer<br>[30:18] scholarship or opportunities.<br>[30:21] Uh I don't know about a specific one but<br>[30:24] for sure there are I would really push<br>[30:27] in something that is related to my<br>[30:29] domain because if you have a domain in<br>[30:31] medicine and economics you can I think<br>[30:33] you I would push in that direction. So<br>[30:35] I'll put in that I don't have a specific<br>[30:37] thing. Yeah. So sorry, Franchesco. There<br>[30:40] was somebody standing right in front of<br>[30:42] my window. So he was not going to stop.<br>[30:45] He was like, "Hello, mom. Someone is<br>[30:48] trying to break into our house right<br>[30:49] now." Like somebody they see me. And I<br>[30:51] signal like I'm see I can come to the<br>[30:53] door and they're like tapping on the<br>[30:55] window. [laughter]<br>[30:56] >> So anyways, uh I'm sorry about that. Uh<br>[30:58] but yes, let's uh let's resume. I I I'm<br>[31:03] trying to think of my train of thought.<br>[31:05] We're talking about using AI. Yes.<br>[31:07] couple of months ago this project that I<br>[31:08] had online um for there they created<br>[31:11] this extension because they were looking<br>[31:12] for jobs and you know how you apply to<br>[31:14] like a million jobs and then you can<br>[31:16] keep track on your email of the<br>[31:18] applications you sent who reply who<br>[31:21] actually like set up an interview with<br>[31:24] you but they created this open source<br>[31:26] project that does that and it's called<br>[31:27] just another app I'll find the link and<br>[31:29] share it uh but the reason I I thought<br>[31:32] about this it's um because while we were<br>[31:35] doing the the the show we were looking<br>[31:38] through like the different roles that<br>[31:40] are open out there and some of the roles<br>[31:42] are for companies that are basically AI<br>[31:45] AI companies like Antropic is it was<br>[31:47] hiring at the time for a ton of roles.<br>[31:49] They're still hiring for a ton of roles<br>[31:50] and I love how they break it down<br>[31:54] what how could they use AI as part of<br>[31:57] your application process because they're<br>[31:59] being realistic about it. You're not<br>[32:01] going to not use AI, right? Like if I if<br>[32:03] I'm trying to tailor my resume to a<br>[32:05] particular role, like realistically, I'm<br>[32:07] going to probably use AI to help me at<br>[32:09] least dissect and and do it faster than<br>[32:12] I would. So, they have a really cool<br>[32:15] guide on their website uh when you're<br>[32:17] looking through their career page that<br>[32:18] actually tells you like this is what a<br>[32:21] good usage of AI is. Basically, it's not<br>[32:23] saying don't use it as a point of of of<br>[32:26] a starting point or point of reference,<br>[32:28] but please don't create your resume<br>[32:30] entirely with AI because it is obvious.<br>[32:32] Like you can 100% tell when when<br>[32:34] something has been literally a copy and<br>[32:36] paste it from uh just chat GPT. Uh, you<br>[32:40] know, I've seen things where people<br>[32:42] forget to remove the like the either the<br>[32:45] footer or the top and then it's like<br>[32:47] it's painful because I bet you actually<br>[32:50] maybe sometimes you even worked a lot on<br>[32:52] that. Uh, but then when it was time to<br>[32:54] send that, you just hit copy and paste<br>[32:55] and there you go. Now I know that like I<br>[32:57] feel like it was less effort effort<br>[32:58] wasted. So I think there is ways for you<br>[33:00] to use AI in a way that is for your<br>[33:02] advantage. I definitely uplevel my<br>[33:05] skills, uh, my technical skills on<br>[33:07] things that I was not super strong on<br>[33:09] with the use of AI. You know, having the<br>[33:11] conversations with the assistant more<br>[33:13] than just letting it write everything<br>[33:15] for me. Um, it's it's helped me level up<br>[33:18] in a lot of just things that were not my<br>[33:20] feel or a muscle that I wasn't flexing<br>[33:22] all the time. Like I don't do front-end<br>[33:24] design. Like thank you for complimenting<br>[33:26] the website. like my first iteration of<br>[33:28] the website was [laughter] not great<br>[33:31] because I am not I'm not a front-end<br>[33:33] person. Um so I've been able to use AI<br>[33:36] to just sort of at least level up on<br>[33:39] those skills. That's one good usage. So<br>[33:41] if you're learning Rust and you're like<br>[33:43] using this uh coding assistance like use<br>[33:46] them more as as that use them more as uh<br>[33:49] leveraging for you to kind of be be a<br>[33:51] conversation and for you to fill up the<br>[33:53] gaps and skills and then when it's time<br>[33:55] to actually apply for a job yes you can<br>[33:56] use it to uh sort of validate your<br>[34:00] resume your experience but adapt it but<br>[34:03] not as a copy and paste situation. And<br>[34:05] then if you're using AI, um I think<br>[34:07] we're missing that that's actually a<br>[34:09] good opportunity for you to put that in<br>[34:11] a resume in a way that is in a positive<br>[34:14] light like not as a negative. Um because<br>[34:17] I'm not going to I mean I would be<br>[34:19] hardpressed to look at a at a resume now<br>[34:22] this today in 2025 and if there's<br>[34:26] someone in the chat right now who's<br>[34:28] never used any kind of AI for coding<br>[34:30] please let me know because I I love to<br>[34:31] meet this unicorn because I'm like no<br>[34:33] there's that like one way or another<br>[34:35] even if you don't know it there's been<br>[34:37] somewhere in automation there has been<br>[34:39] somewhere powered by AI that you've used<br>[34:41] so own that narrative right like frame<br>[34:44] it more as a it was a tool that I use to<br>[34:47] verify the judgment that I was making or<br>[34:50] help inform a decision or help me expand<br>[34:53] the view on something like I'm telling<br>[34:55] you right now these are things I'm not<br>[34:56] skilled at. So I am going to sort of use<br>[34:59] it as a rubber ducky and kind of help me<br>[35:02] expand my knowledge there. And so if I<br>[35:04] need to list that on a resume, I'm going<br>[35:06] to list that on a resume, but I'm not<br>[35:07] going to write like I use GitHub Copilot<br>[35:10] and that's the bullet point on my<br>[35:11] resume. Um that's like saying like I use<br>[35:14] Google [laughter] everybody does it but<br>[35:16] it's more like okay well I use GitHub<br>[35:18] copilot uh and AI assistant development<br>[35:22] focus on the design of my page and I<br>[35:28] used it leveraging my architecture<br>[35:30] decisions that I had made for my page<br>[35:31] and this is the specifications that I<br>[35:33] created and using AI assistant<br>[35:37] development I was able to I don't know<br>[35:39] cut the production time on X Y or that<br>[35:42] um or I created these validation<br>[35:45] workflows for this AI generated code in<br>[35:49] production. So there are ways that you<br>[35:50] can leverage the things that you're<br>[35:51] doing with AI in a positive light and a<br>[35:54] light that actually shows a a<br>[35:56] prospective employer that yes, you're a<br>[35:59] person that is you know your staff, you<br>[36:01] know your fundamentals, but you're also<br>[36:04] open and eager to work with the<br>[36:05] technologies that I think every company<br>[36:07] right now has a mandate to use AI in one<br>[36:09] way or the other, especially if you're a<br>[36:11] public company cuz investors love to be<br>[36:13] with the buzzy trends and the you know<br>[36:16] that that's what's hype so that's what<br>[36:17] they want to hype. Um, so you know, like<br>[36:20] there's a way that you can position your<br>[36:21] usage of it in a positive light. Uh, and<br>[36:24] then also like use it as a as a tool for<br>[36:27] you to learn. Um, not to replace your<br>[36:31] fundamental learning, your fundamental<br>[36:33] practices. Like I've been poking around<br>[36:37] trying to do some things with Rust for<br>[36:39] ever and ever. and and I've had I have<br>[36:42] some really smart people, shout out to<br>[36:44] Miguel Ras, who actually tried to help<br>[36:46] me. Uh, but not consistently. I I I was<br>[36:49] just like, you know what? I I just don't<br>[36:50] I don't I don't get it. Like, and I've<br>[36:52] done the same Robert Ducking thing that<br>[36:54] I would do with Copilot, trying to look<br>[36:56] at some skilled things on Rust, but I<br>[36:59] watched your boot camp and a lot more<br>[37:01] things clicked because it was like<br>[37:03] actual dedicated time for learning. So,<br>[37:07] you know, yes, it's good to use it in in<br>[37:10] in a way that you're experimenting and<br>[37:12] kind of stretching yourself out there.<br>[37:14] Lean into that AI because it's always<br>[37:16] there. You know, that's that's when you<br>[37:17] have time and you can I can quickly have<br>[37:19] a conversation about like review this<br>[37:21] diagram and tell me how I could do this<br>[37:22] better or I do a lot of security things<br>[37:24] with AI too. like I but with params that<br>[37:27] I've collected with my own<br>[37:29] specifications of what I know to be good<br>[37:31] security practices then I feed that into<br>[37:34] it and say okay review this and tell me<br>[37:36] where there are holes what could I do<br>[37:38] better how can I improve this and it<br>[37:40] it's just a way to leverageing it um not<br>[37:42] replacing of course that fact that I<br>[37:44] needed to have that CSK I needed to know<br>[37:46] those fundamentals in order to know what<br>[37:48] I'm asking same thing with learning Rust<br>[37:50] I'm like oh I finally got a lot more<br>[37:53] just by actually watching an educational<br>[37:55] thing, but now moving forward like I can<br>[37:57] actually apply those learnings into the<br>[37:59] way that I'm I'm I'm using the tools. So<br>[38:01] I I wouldn't shy away from from using it<br>[38:03] and experimenting and using it to grow.<br>[38:06] But please please don't forget the<br>[38:08] fundamentals are so so important. And I<br>[38:10] think the the persons that are coming<br>[38:12] into the industry that have that solid<br>[38:14] understanding of fundamentals and also<br>[38:16] the flexibility and eagerness to use the<br>[38:19] tools are the ones that are going to<br>[38:21] succeed long term.<br>[38:25] Yeah, I've been saying fundamentals are<br>[38:27] fundamental I think for the last 20<br>[38:29] years and the more the year they pass<br>[38:32] the more it becomes more important. By<br>[38:35] the way, I'm not I'm not crying tears of<br>[38:37] joy, but I have something in my eye. I'm<br>[38:39] sorry, Andrea, but I don't know. Maybe I<br>[38:41] stared at my screen for too long. So,<br>[38:43] I'm not it's not it's not crying tears<br>[38:45] of joy for what you said. [laughter]<br>[38:48] Um, yes. Thank you. I would I would like<br>[38:52] to you to try to say something about the<br>[38:56] specific about GitHub because of course<br>[38:58] you work there as developer advocate. Is<br>[39:01] there anything like in GitHub that can<br>[39:04] help you? If you have some some some<br>[39:07] examples that maybe are not known by<br>[39:09] many people but like how you can uh you<br>[39:12] know is there any GitHub feature? Maybe<br>[39:15] you have something that comes up not on<br>[39:17] top of your mind. You're muted<br>[39:19] >> for [snorts] you to get higher or like<br>[39:21] if you're looking for a job or you're<br>[39:22] looking to learn.<br>[39:23] >> Yes. Let's say maybe a feature that not<br>[39:25] many people know about GitHub that can<br>[39:28] help them in their career. Let's say<br>[39:30] whatever comes up in your mind if you<br>[39:32] have anything.<br>[39:33] >> Oh, I got you. Okay. Um, so<br>[39:38] for your career visibility for sure, I<br>[39:40] would say the same things the same<br>[39:42] points that we touched earlier. have a<br>[39:44] you know it is important to have a<br>[39:47] profile read me. Um a lot of people<br>[39:50] either leave it blank or they just write<br>[39:51] something like hi I'm Franchesco and I<br>[39:54] like coding.<br>[39:56] Okay that doesn't tell me a lot about<br>[39:58] Franchesco. So definitely taking the<br>[39:59] time to kind of stand out there to show<br>[40:01] what you're currently working on. uh<br>[40:04] when you show what you're working on<br>[40:05] that's like your you're exciting like<br>[40:08] that it it comes through like I can I<br>[40:10] feel like if I feel your enthusiasm for<br>[40:12] whatever it is like I know you're<br>[40:14] enthusiastic how Russ<br>[40:18] um so sort of bringing that same energy<br>[40:21] to your read me is something that is a<br>[40:23] underestimated thing. Um, and then<br>[40:25] making it super clear on what you're<br>[40:27] looking for. That's a good space for you<br>[40:29] to do that, right? Um, that's a place<br>[40:31] where you can literally write whatever<br>[40:32] you want. So, you're like, I'm seeking<br>[40:33] for a role that will help me, you know,<br>[40:36] or I have this amount of amazing skill.<br>[40:39] I know I can be impactful in this type<br>[40:41] of place. Uh, that's is your read me.<br>[40:44] You can write whatever you want there.<br>[40:46] So uh anything that you have that is<br>[40:49] outwards like we use a lot of link trees<br>[40:52] and different linking and if you have a<br>[40:53] personal website that's great. Uh but<br>[40:56] also this is a space where you can link<br>[40:57] everything. So if if you've been to a<br>[40:59] podcast with Franchesco put it there. If<br>[41:02] you are speaking at a conference link<br>[41:05] the conference and also at the same time<br>[41:07] you also helping increase visibility for<br>[41:09] that conference. So use that profile<br>[41:11] readme space. It's a good piece of real<br>[41:14] estate where you can literally do<br>[41:16] anything that you want. Uh, and again<br>[41:18] the first thing that people see when<br>[41:20] they land on your GitHub page. So I<br>[41:21] think that's a underrated underrated<br>[41:23] underused skill. One thing that maybe is<br>[41:25] not as mainstream and maybe recently<br>[41:29] because I've been doing a lot of work in<br>[41:31] this particular in this particular<br>[41:34] point. Uh, it's something that I've been<br>[41:36] looking at niche for learning for<br>[41:39] upleveling my own skill. But it's also<br>[41:41] kind of made me think about, huh, this<br>[41:43] is an interesting way to see how people<br>[41:45] approach uh CI/CD pipelines. I look at<br>[41:48] the GitHub actions workflow tabs in your<br>[41:50] repositories. I do. I do. I do because<br>[41:54] I've been in that YAML I'm going to stop<br>[41:56] hating on YAML and start calling it YAML<br>[41:58] hell because I feel like I'm setting<br>[41:59] myself up for failure there. This is<br>[42:01] where we are. So what we got? [laughter]<br>[42:04] So we're I'm going to say from now on I<br>[42:05] love YAML and I've been doing a lot of<br>[42:08] things that have required me to look<br>[42:10] deeper into the way that folks are<br>[42:11] integrating this. And so I'm looking I'm<br>[42:14] actually looking through your action<br>[42:15] stuff and it does kind of made me think<br>[42:18] about a different angle of looking at<br>[42:21] how you think about quality um<br>[42:26] how you think about optimization.<br>[42:28] It's interesting to me to to parse<br>[42:30] through that. So believe it or not,<br>[42:31] there are nosy people out there that are<br>[42:33] looking at things like that. [laughter]<br>[42:34] So you know those those are just two<br>[42:36] things that you can do right away in<br>[42:38] GitHub. Um there is a very little known<br>[42:41] place I hope to be known by more people<br>[42:43] actually like our work in the team<br>[42:45] before it was what it is now. You know<br>[42:47] we have a discussions board in GitHub<br>[42:49] like literally the GitHub discussions<br>[42:51] that is for the entire company that<br>[42:53] started as a discourse instance many<br>[42:56] many years ago that's when I worked on<br>[42:57] it. uh but it's now integrated part of<br>[42:59] the platform. So if you are thinking hey<br>[43:02] you know what I want to go learn<br>[43:03] something new or I want to know share or<br>[43:05] I want to have a conversation about a<br>[43:06] GitHub feature um go have it in public<br>[43:10] go learn in public with with there's a<br>[43:12] lot of great people there. That's also a<br>[43:13] good place to poke uh product managers<br>[43:16] and engineering folks from GitHub<br>[43:17] because they do spend time there and<br>[43:19] they do read the comments. Uh I know a<br>[43:21] lot of people love Reddit and all that<br>[43:23] but I'm like I'm not going somewhere<br>[43:24] else. My code is on GitHub. my<br>[43:27] discussions about my code and GitHub.<br>[43:29] Uh, and if if you're if you're looking<br>[43:31] for a place where you can actually help<br>[43:33] out people and show what you know,<br>[43:35] that's a good place for it. I can share<br>[43:37] the link of it as well. Um, so that's<br>[43:39] another also underrated place. I know<br>[43:41] people that got in jobs because they<br>[43:44] have like they apply for companies uh<br>[43:47] maybe about a specific part of I don't<br>[43:49] know GitHub as a platform or maybe about<br>[43:53] a specific um maybe feature or<br>[43:59] and under like this if I'm looking for<br>[44:01] someone to come and optimize my CI/CD<br>[44:03] pipelines like I'm gonna look at your<br>[44:05] actions on your own on your own projects<br>[44:09] but also I might look I'm also O going<br>[44:11] to look at your discussions and see if<br>[44:14] you have actually if you have had these<br>[44:15] conversations in the open and I'll share<br>[44:17] I'm sharing the links with you now on<br>[44:19] the private uh site that's the link for<br>[44:21] the discussions um and then I'll share<br>[44:24] the the link for the uh just a job app<br>[44:27] because it's a great project and<br>[44:29] everyone who's looking for a job should<br>[44:30] definitely leverage it um so yeah so<br>[44:33] it's it's uh it's I'm going to look<br>[44:35] through that um<br>[44:38] >> everything that you're doing in the open<br>[44:40] it tells me a lot more about how you're<br>[44:41] going to work in my team and that's<br>[44:44] maybe an underrated type of thing. Uh<br>[44:47] and honestly there is a lot of spaces<br>[44:49] now for you to kind of voice opinions<br>[44:51] and I would I would recommend keeping<br>[44:54] them in there as well because there is<br>[44:56] it's one less place that your recruiter<br>[44:58] has to go find. [laughter]<br>[45:02] >> Perfect. Thank you. And uh so watch out<br>[45:06] be careful because everything you do<br>[45:07] online will be seen in some ways. This<br>[45:10] is not just GitHub but let's say for<br>[45:12] GitHub it's a the professional uh place<br>[45:16] to where you can also add you know a bit<br>[45:19] of your personality of what what who you<br>[45:21] are. It's a mix between I like what you<br>[45:24] said like not just write I I I I write<br>[45:27] code. Of course, we are giving these<br>[45:29] tips if you're just getting started.<br>[45:31] Once you have on have been working in<br>[45:32] the industry for 20 years, probably you<br>[45:34] don't need to mention everything. We are<br>[45:37] trying to optimize these tips for who is<br>[45:39] just getting started. They are<br>[45:40] struggling for finding a job once the<br>[45:43] world knows you. I don't I don't think<br>[45:44] that Leno store needs to explain who he<br>[45:47] is. He doesn't need anymore. It's the<br>[45:49] same thing when we are on social media.<br>[45:52] Once you have a million followers on<br>[45:54] YouTube, your Twitter header, nobody<br>[45:56] could could care less. But if you are<br>[45:58] just at the beginning, it might help you<br>[46:00] in getting a bit noticed in what you do.<br>[46:02] So I really like that your your approach<br>[46:05] because it is for for the be for the<br>[46:08] beginner. The internet is forever. You<br>[46:10] are right. And now I think everyone is<br>[46:13] checking their own GitHub profile. I can<br>[46:14] check it because now I'm live, but I<br>[46:16] will check it of course later because I<br>[46:19] think that it's important to check it<br>[46:21] once once in a while. Andre, I don't<br>[46:24] know if when you have to go. I will just<br>[46:26] before you leave, I would like just to<br>[46:28] ask you what which technologies and<br>[46:31] frameworks I think you heard are the<br>[46:34] things to bet on in 2026. We are at the<br>[46:37] end of 2025. What do you think are the<br>[46:39] technologies that developers should bet<br>[46:42] on if we are betting our time as money?<br>[46:46] [snorts]<br>[46:48] >> Francisco, are you trying to get me in<br>[46:49] trouble here? Like no<br>[46:51] >> technologies technologies are like<br>[46:53] babies like you can't you can't<br>[46:54] [laughter]<br>[46:55] I can't say I have a favorite one uh<br>[46:58] somebody in the chat say definitely not<br>[46:59] Russ [laughter]<br>[47:03] or<br>[47:05] >> I love it I love it okay like so we do<br>[47:07] this report every year at GitHub it's<br>[47:09] called the state of the octoverse report<br>[47:11] I can share a link to it as well um to<br>[47:14] you so that you can share on the chat<br>[47:16] but it's a really good report in that it<br>[47:18] gives you a sort of highle overview of<br>[47:20] the what we know and and if you think<br>[47:22] about what GitHub knows like we have the<br>[47:24] power of data. Um so it tells you you<br>[47:28] know like the the growing technologies<br>[47:30] the what programming languages are like<br>[47:33] the most used in GitHub what projects uh<br>[47:37] or communities or even parts of the<br>[47:39] world sold the biggest growth. So there<br>[47:42] is a lot there. Um, I'm not saying go<br>[47:46] look at this report and use it as your<br>[47:48] like bible like everything that oh it<br>[47:51] says TypeScript just surpass Python so<br>[47:54] that's it. I'm going to be full on<br>[47:56] TypeScript. No. Um, I think it does give<br>[47:59] you a good show of like the industry as<br>[48:01] a whole of what's growing, what's sort<br>[48:04] of what's what's what's changing. Um, I<br>[48:08] think right now we're seeing a lot of<br>[48:10] growth that's driven by frameworks that<br>[48:13] are doing things like scaffolding<br>[48:14] TypeScript by default because AI and AI<br>[48:17] assistant development that benefits from<br>[48:19] that. Uh, but there is also some<br>[48:21] interesting finds in that report like<br>[48:23] there is there was like a tiny tiniest<br>[48:26] meion of a race on the uh usage of<br>[48:29] cobalt which is such an old it's like 60<br>[48:32] is it 65 years old? It might be. Yeah, I<br>[48:35] think it might be 65 years old. Um, it's<br>[48:38] a very old very old programming<br>[48:40] language. Uh, but there is like there<br>[48:43] was a growth on the active developers<br>[48:45] that are actually using that technology<br>[48:48] because now they're with the help of AI<br>[48:51] we're looking at modernization of some<br>[48:52] of the things maybe things that weren't<br>[48:54] touched from years and years ago because<br>[48:56] that's the way that it work and the<br>[48:58] people who were working on it alone<br>[49:00] retire. But now with the help of AI,<br>[49:02] we're actually being more realistic<br>[49:04] about what can be modernized. So it's<br>[49:06] kind of interesting. I think I wouldn't<br>[49:09] pin it so much on one particular<br>[49:12] language or one particular framework.<br>[49:15] Although I got to say I've been looking<br>[49:17] at Tanstack and it's so good, y'all.<br>[49:19] Like I was looking at a comparison.<br>[49:21] Somebody wrote a really good article of<br>[49:23] a comparison of that in Nex.js. Um and<br>[49:26] it's I I mean it's that team is cooking.<br>[49:28] They're doing a lot of great things. I<br>[49:30] think a lot of it is going to be more I<br>[49:33] wouldn't focus so much on one particular<br>[49:36] programming language or one particular<br>[49:38] framework. Um if you want to look at one<br>[49:41] interesting space and you're thinking of<br>[49:43] like maybe I haven't decided what I want<br>[49:45] to do. Uh one thing that we can<br>[49:47] definitely distill from that October<br>[49:50] report is the growth of infrastructure<br>[49:53] as code practices. That's that's the way<br>[49:56] the the the world is spinning now. Uh so<br>[49:59] that might be an interesting space for<br>[50:01] you to take a look at if you've if you<br>[50:03] never actually kind of dug into that<br>[50:06] before. So I wouldn't pin it so much<br>[50:08] onto like oh this is the upcoming<br>[50:10] language although hey shout out to all<br>[50:11] of you crop stations out there like the<br>[50:13] ROS community is also cooking. Uh but<br>[50:16] something like infrastructure or code as<br>[50:18] code uh systems programming if you want<br>[50:21] to be ahead of the tool may ahead of the<br>[50:23] the curve maybe looking more at the not<br>[50:26] so much the things that are like super<br>[50:28] shiny and new and and glamorous but more<br>[50:30] the things that are actually in<br>[50:31] production and cooking like uh I so I<br>[50:34] can't say one particular language or<br>[50:36] framework I would say more look at the<br>[50:38] technologies as a whole and then the<br>[50:41] sleeper the sleeper out there looking at<br>[50:43] at the systems design spec<br>[50:45] and infrastructure as code practices<br>[50:48] because those are the things that are<br>[50:49] that have seen the most significant<br>[50:51] growth and that's not going to that's<br>[50:53] not going to change like that's only<br>[50:54] going to keep shooting upwards.<br>[50:58] >> Perfect. Thank you. Thank you Andrea.<br>[51:00] And something about unstuck might come<br>[51:02] up soon on this channel but I'm not<br>[51:04] seeing a single more word on it. I'm not<br>[51:08] seeing anything of it, but let me cook<br>[51:11] and let me do my own stuff when I'm not<br>[51:13] recording uh podcast. Of course, going<br>[51:16] live.<br>[51:17] >> Uh Andrea, I don't know if you have to<br>[51:18] go if you have uh something coming up<br>[51:21] now or or not.<br>[51:23] >> I got I got like five minutes. I So,<br>[51:25] friends, for you who speak Spanish,<br>[51:27] please join me on the quack which will<br>[51:30] >> drop the drop the let me drop<br>[51:33] >> drop the link.<br>[51:34] >> Drop the link. Ordev lives in Spain. So<br>[51:37] it might might actually join. [laughter]<br>[51:40] >> Yeah. Oh yeah, there you go. I mean you<br>[51:43] can practice your your Spanish and I<br>[51:44] don't know if you have a big Italian<br>[51:45] follow in Italian and Spanish is very<br>[51:47] similar.<br>[51:48] >> Well I there is this stereotype that<br>[51:50] Italians can understand Spanish and it's<br>[51:53] quite true. Like I never studied Spanish<br>[51:55] but I still can understand 75% of the<br>[51:58] Spanish.<br>[51:59] >> Same [snorts] same.<br>[52:01] >> Sort of.<br>[52:03] >> Sort of. You like you sent me the link<br>[52:05] here. Perfect.<br>[52:08] >> Okay. Also, I am I'm recording a podcast<br>[52:11] after this. I recorded a podcast before<br>[52:13] that before this live stream. So, we're<br>[52:16] always busy, but um I really really<br>[52:19] enjoyed this conversation, Andrea. And<br>[52:23] >> thank you, Franchesco. Thank you so much<br>[52:25] for having me. Like, I this was fun.<br>[52:28] >> Uh please, friends, if you're watching<br>[52:30] this now or after the fact, like reach<br>[52:34] out. like um my DMs are open for real<br>[52:36] and if there is anything that I can help<br>[52:38] you with like I'm happy to try. Um and<br>[52:42] come come to the next come to all the<br>[52:44] things like we'll have this we do on<br>[52:46] Thursdays we do streams uh later on<br>[52:48] Cassidy will be live uh with the rubber<br>[52:51] duck Thursday stream.<br>[52:53] >> Uh I will be doing the rubber duck<br>[52:54] Thursday stream but in Spanish coming up<br>[52:57] next and then of course on Fridays is<br>[52:59] open source Friday every Friday. So come<br>[53:02] come support open source. Perfect. And<br>[53:04] thank you so much. Of course, they were<br>[53:05] inviting me, I think, two times. Yeah.<br>[53:08] >> Yes. Yes. Please. Oh, and please, please<br>[53:10] read this thing that I'm writing so that<br>[53:12] I keep writing it<br>[53:13] >> because I I I wanted to do something<br>[53:16] that was<br>[53:17] >> um that could be fully opinionated and<br>[53:20] not have um to go through the layers of<br>[53:24] things that have to you your friends. I<br>[53:26] mean, I'm sure you know, you wouldn't<br>[53:27] believe the amount of layers of things<br>[53:29] have to go through before they're hit<br>[53:31] the out public when you work for a<br>[53:33] company. So, I'm I'm writing my own<br>[53:35] newsletter now and I'd love for all of<br>[53:36] you to give it a look and give me some<br>[53:37] feedback on it. That would be great.<br>[53:39] >> Perfect. We'll leave the links and<br>[53:41] Andrea, thanks again for being here.<br>[53:44] Thank you so much for everyone who came<br>[53:46] here in this I think it has been a while<br>[53:48] since I haven't done like a live podcast<br>[53:51] but I've been quite busy with<br>[53:52] conferences but I really wanted to have<br>[53:55] Andrea before 2025 ended and I achieved<br>[53:58] my goal so Andrea we made it. Thank you.<br>[54:01] This made me very<br>[54:04] thank you so much.<br>[54:06] >> I appreciate you. Thank you so much<br>[54:08] everyone who joined. Um sorry I was I<br>[54:10] wasn't able to like reply on the chat<br>[54:12] but I appreciate so much. It's fine. See<br>[54:15] you the next time. Bye-bye.<br>[54:16] >> Bye.<br>[54:18] >> Uh, where is the button? Here.<br>[54:22] No, here.<br>[54:25] It's already live. How do I stop it?<br>[54:30] Interesting. See you all. I think it's<br>[54:34] bugged. Let me check. How do I <!-- deep-read:sections end --> ## ๐Ÿ“บ Video ![](https://www.youtube.com/watch?v=I9zQos_ePDs) --- # ๐Ÿ“ Full Notes (Deep-Read Analysis) # ๐Ÿท๏ธ Title: Dev Career Secrets from a GitHub Strategy Leader ## ๐Ÿงฐ Action Inbox - [H1] Update GitHub profile README with specific technologies and personality โ€” 10 seconds recruiter scan #career - [H2] Pin repositories relevant to target role, not everything โ€” quality over quantity showcase #portfolio - [H3] Document learning journey publicly on GitHub discussions โ€” build community reputation #networking ## ๐Ÿงญ Table of Contents - [๐Ÿ“ Summary](#-summary) - [๐Ÿ“– Main Content](#-main-content) - [๐Ÿ—’๏ธ Notes](#๏ธ-notes) - [๐Ÿ“š References](#-references) - [๐Ÿ—ฃ๏ธ Paraphrase](#๏ธ-paraphrase) - [๐Ÿ“ฐ Newsletter](#-newsletter) - [โœ… Action Items](#-action-items) - [๐Ÿงช Hypotheses](#-hypotheses) - [๐Ÿ“Š Methodology](#-methodology) - [โ“ Questions](#-questions) - [๐Ÿ”ฎ Future Research](#-future-research) - [๐Ÿ’ก Implications](#-implications) - [๐Ÿง  Learning Styles Assessment](#-learning-styles-assessment) - [๐Ÿ“Š Accessibility Assessment](#-accessibility-assessment) - [๐ŸŒ DEIB Assessment](#-deib-assessment) - [โœ… Logical Validity](#-logical-validity) - [๐Ÿ›ก๏ธ DARVO Analysis](#-darvo-analysis) - [๐ŸŽฏ Conclusion](#-conclusion) ### ๐Ÿ”— References - [GitHub State of the Octoverse Report](https://github.com) - [GitHub Discussions](https://github.com) - [Just Another Job Seeking App (JAJA)](https://github.com) ## ๐Ÿ“ Summary - **What happened**: GitHub developer advocate Andrea Griffiths discusses developer career strategies for 2026, addressing layoffs, junior developer challenges, and effective use of AI in job seeking - **Why it matters**: Tech job market has become more competitive with higher barriers to entry, requiring new strategies for career success - **Key stakeholders**: Junior developers, career changers, job seekers in tech - **Risks and tradeoffs**: AI can help learning but over-reliance damages authenticity; networking helps but requires genuine engagement - **Bottom line**: Junior developer roles haven't disappeared but evolved - success requires deeper fundamentals, strategic GitHub presence, and authentic community participation ## ๐Ÿ“– Main Content Andrea Griffiths, a career changer turned GitHub developer advocate, explains how the tech hiring landscape has shifted. Junior developer roles now require broader system thinking beyond basic coding - AI has raised the bar by making simple coding accessible to everyone. Success depends on understanding debugging, security, architecture, and business context. The conversation addresses common misconceptions about GitHub contribution graphs (green squares don't equal hiring potential) and emphasizes strategic profile optimization. Recruiters spend only 10 seconds on initial profile scans, focusing on pinned repositories, bio information, and recent activity quality over quantity. AI should be leveraged as a learning tool and conversation partner, not a replacement for fundamental skills. Companies expect AI usage but value developers who can architect, debug, and make security decisions. Building authentic community presence through documentation, open source contributions, and public learning creates more opportunities than traditional job applications. ## ๐Ÿ—’๏ธ Notes #### โฑ๏ธ [00:00](https://youtu.be/I9zQos_ePDs?t=0) โ€” Introduction and context setting - Key idea: Discussing developer careers for 2026 amid challenging job market - Evidence: "It has been a tough year for most of us layoffs" #### โฑ๏ธ [03:45](https://youtu.be/I9zQos_ePDs?t=225) โ€” Andrea's non-traditional background - Key idea: Career changer perspective on tech industry evolution - Evidence: "I'm a career changer... been doing this now for almost 10 years" #### โฑ๏ธ [06:30](https://youtu.be/I9zQos_ePDs?t=390) โ€” Junior developer role transformation - Key idea: AI has reshaped entry-level expectations, not eliminated roles - Evidence: "No one is born a senior engineer... but now there is a higher bar" #### โฑ๏ธ [09:15](https://youtu.be/I9zQos_ePDs?t=555) โ€” Beyond basic coding skills - Key idea: Modern juniors need architecture and systems thinking from day one - Evidence: "Can you debug that app? Can you put proper security measures in place?" #### โฑ๏ธ [12:00](https://youtu.be/I9zQos_ePDs?t=720) โ€” Green squares misconception - Key idea: GitHub contribution graphs don't indicate developer quality - Evidence: "I don't judge a person's capability by how many green squares" #### โฑ๏ธ [15:20](https://youtu.be/I9zQos_ePDs?t=920) โ€” 10-second recruiter scan strategy - Key idea: First impression depends on pinned repos, bio, and recent activity - Evidence: "Your pin repos... what are you choosing to show me?" #### โฑ๏ธ [18:45](https://youtu.be/I9zQos_ePDs?t=1125) โ€” Making yourself hireable - Key idea: Networking and community participation outweigh cold applications - Evidence: "Nine times out of 10, you're not even really applying for jobs" #### โฑ๏ธ [22:10](https://youtu.be/I9zQos_ePDs?t=1330) โ€” Documentation and storytelling - Key idea: Public learning journeys create opportunities and connections - Evidence: "Humans we love to follow a story... document everything that you're doing" #### โฑ๏ธ [26:30](https://youtu.be/I9zQos_ePDs?t=1590) โ€” Authentic online presence - Key idea: Have opinions and engage genuinely, follower count doesn't matter - Evidence: "You never know who's actually looking at that and who's reading" #### โฑ๏ธ [30:15](https://youtu.be/I9zQos_ePDs?t=1815) โ€” AI as learning tool, not replacement - Key idea: Use AI for conversation and skill gaps, not resume generation - Evidence: "Use them more as leveraging for you to kind of fill up the gaps" #### โฑ๏ธ [35:00](https://youtu.be/I9zQos_ePDs?t=2100) โ€” Positioning AI usage positively - Key idea: Frame AI experience as architectural decision-making enhancement - Evidence: "I used it leveraging my architecture decisions... cut the production time" #### โฑ๏ธ [38:40](https://youtu.be/I9zQos_ePDs?t=2320) โ€” GitHub features for career growth - Key idea: Profile README, GitHub Actions, and Discussions board optimization - Evidence: "I look at the GitHub actions workflow tabs in your repositories" #### โฑ๏ธ [42:20](https://youtu.be/I9zQos_ePDs?t=2540) โ€” Technology trends for 2026 - Key idea: Focus on infrastructure as code and systems programming - Evidence: "Growth of infrastructure as code practices... that's the way the world is spinning" ## ๐Ÿ“š References - [GitHub State of the Octoverse Report](https://github.com) - [GitHub Discussions Community](https://github.com/discussions) - [Just Another Job Seeking App (JAJA) - Open Source Project](https://github.com) - [Anthropic Career Guidelines](https://anthropic.com/careers) ## ๐Ÿ—ฃ๏ธ Paraphrase - Junior developer roles have evolved rather than disappeared - AI has raised the entry bar by making basic coding accessible to everyone, so new developers need broader systems thinking including debugging, security, and architecture skills from the start - Strategic GitHub profile optimization matters more than activity volume - recruiters scan profiles in 10 seconds focusing on pinned repositories, bio clarity, and meaningful contribution patterns rather than green square quantities - Effective AI usage involves leveraging it as a learning conversation partner and architectural decision aid while maintaining fundamental skills and framing AI experience positively on resumes rather than hiding or over-depending on it ## ๐Ÿ“ฐ Newsletter Ready to level up your developer career in 2026? ๐Ÿš€ GitHub Developer Advocate Andrea Griffiths shares game-changing insights about the evolving tech job market. The key takeaway? Junior developer roles aren't disappearing - they're transforming. AI has raised the bar, making basic coding skills table stakes while demanding deeper understanding of systems, security, and architecture. **Your career action plan:** - Optimize your GitHub profile strategically (pinned repos > green squares) - Use AI as a learning partner, not a replacement for fundamentals - Build authentic community presence through public documentation - Focus on infrastructure as code and systems programming trends The WordPress community can especially benefit from these insights as we navigate our own ecosystem changes. Consider how these strategies apply to contributing to Core, building plugins, or advancing your WordPress development career. Resources: [GitHub Discussions](https://github.com/discussions), [State of the Octoverse](https://github.blog), [Learn WordPress](https://learn.wordpress.org/), [Developer Docs](https://developer.wordpress.org/) ## โœ… Action Items #### DevRel - Developer advocacy: Create GitHub profile optimization workshop series with live examples - Education: Host office hours on AI usage best practices for career development - Developer experience: Document GitHub features for career advancement (Actions, Discussions) - Community management: Moderate discussions around career advice and job market trends #### WordPress - Core: Encourage meaningful GitHub activity patterns for contributors - Community: Create mentorship programs pairing experienced developers with career changers - Training: Develop lesson plans on systems thinking and architecture for junior developers - Documentation: Style guides for professional GitHub presence and portfolio development #### Community - Maintainers: Support authentic contribution recognition beyond commit counts - Contributors: Onboarding programs emphasizing public learning and documentation - Sponsors: Fund career development initiatives for underrepresented developers ## ๐Ÿงช Hypotheses - Developers who document their learning journey publicly have higher hiring success rates than those who code privately - GitHub profiles with strategically curated pinned repositories receive more recruiter attention than profiles with numerous unorganized repos - Career changers who leverage domain expertise alongside technical skills outperform traditional CS graduates in certain roles - Companies increasingly value AI-assisted development skills when presented as architectural decision-making tools ### ๐Ÿ“Š Methodology - Track hiring outcomes for developers who maintain public learning documentation vs. private portfolios - A/B test GitHub profiles with strategic curation vs. comprehensive showcasing - Survey recruiters on profile scanning patterns and decision factors - Analyze correlation between community participation and career advancement ### โ“ Questions - How do hiring managers differentiate between authentic AI usage and over-reliance during technical interviews? - What specific community contributions create the strongest networking effects for career advancement? - How should career changers balance domain expertise with technical skill development in competitive markets? ### ๐Ÿ”ฎ Future Research - Impact of AI coding assistants on junior developer skill development pathways - Effectiveness of different professional networking strategies in remote-first tech culture - Evolution of technical interview processes as AI tools become standard development practice ### ๐Ÿ’ก Implications - **Practice**: Shift from quantity-focused coding practice to quality-focused systems thinking and documentation - **Education**: Update bootcamp and university curricula to emphasize architecture and debugging alongside basic programming - **Hiring**: Recruiter training on meaningful GitHub activity patterns vs. superficial metrics - **Career Development**: Investment in public learning platforms and community participation tools ### ๐Ÿง  Learning Styles Assessment Reading level estimate: 10th-11th grade (accessible to most developers) #### ๐Ÿ‘€ Visual learners - Present: Speaker video, occasional screen sharing of GitHub profiles - Improve: Diagrams of career pathway progression, visual GitHub optimization examples #### ๐Ÿ”Š Aural learners - Present: Full audio conversation with clear dialogue - Improve: Timestamped audio summaries, speaker notes for key concepts #### ๐Ÿ—ฃ๏ธ Verbal learners - Present: Clear definitions of terms like "infrastructure as code", practical examples - Improve: Career terminology glossary, analogies for technical concepts #### ๐Ÿงฉ Kinesthetic learners - Present: Actionable GitHub profile optimization steps - Improve: Interactive GitHub audit checklists, hands-on portfolio exercises #### ๐Ÿง  Logical learners - Present: Structured argument progression, cause-effect relationships - Improve: Career decision flowcharts, hiring process timelines #### ๐Ÿง‘โ€๐Ÿคโ€๐Ÿง‘ Social learners - Present: Community-focused advice, networking emphasis - Improve: Discussion prompts for developer communities, peer review guidelines #### ๐Ÿง˜โ€โ™€๏ธ Solitary learners - Present: Self-reflection opportunities on career direction - Improve: Personal career assessment worksheets, individual skill development paths ### ๐Ÿ“Š Accessibility Assessment #### Screen reader semantics - Present: Clear speaker transitions, descriptive context - Improve: Transcript with speaker labels, heading structure for easy navigation #### Keyboard navigation - Present: Standard YouTube player controls - Improve: Timestamp links for easy topic jumping #### Contrast and readability - Present: Good video quality with clear speakers - Improve: High contrast transcript option, adjustable text sizing #### Media and captions - Present: Auto-generated captions available - Improve: Professional captions with technical term accuracy, downloadable transcript ### ๐ŸŒ DEIB Assessment #### Cultural breadth - Present: Career changer perspective, non-traditional pathway discussion - Improve: Multiple geographic hiring market examples, international developer experiences #### Inclusive language - Present: Gender-neutral role descriptions, welcoming tone for beginners - Improve: Avoid industry jargon without definition, provide context for technical terms #### Representation - Present: Female developer advocate sharing experiences - Improve: Acknowledge different learning styles, career change motivations #### Psychological safety - Present: Supportive tone about job market challenges, realistic expectations - Improve: Content warnings about layoff discussions, resource links for support ### โœ… Logical Validity #### Strong reasoning - Deductive strength: "AI makes basic coding accessible โ†’ higher entry barriers โ†’ need for systems thinking" [09:15] - Inductive strength: Personal experience and recruiting observations support profile optimization advice [15:20] - Causal reasoning: Clear connection between community participation and hiring opportunities [18:45] #### Fallacies - Appeal to authority: Minimal - relies on experience rather than credentials - Hasty generalization: Some broad claims about "everyone uses AI" lack nuance [30:15] #### Tone of voice - Overall tone: Encouraging and realistic, balances optimism with practical challenges ### ๐Ÿ›ก๏ธ DARVO Analysis - Deny: No defensive patterns observed - Attack: No attacks on accusers or critics - Reverse victim/offender: No victim/offender role reversals - Emotional cues: Genuine empathy for job seekers, authentic career sharing ## ๐ŸŽฏ Conclusion The developer career landscape has fundamentally shifted - junior roles require more comprehensive skills while AI becomes both challenge and opportunity. Success in 2026 depends on strategic GitHub presence, authentic community engagement, and positioning AI as an architectural enhancement tool rather than a replacement for fundamental skills. Focus on systems thinking, public learning documentation, and building genuine professional relationships through community participation. ## ๐Ÿงฑ Refactor Appendix ### ๐Ÿ“š Literature Split Pad Split level: H3 ### [H1] Junior developer roles require systems thinking beyond basic coding > [!quote] Dev Career Secrets from a GitHub Strategy Leader by Andrea Griffiths via YouTube. Published 2024-12-28. > "Can you debug that app? Can you put proper security measures in place?" AI accessibility has raised entry-level expectations requiring architecture knowledge. [view highlight](https://youtu.be/I9zQos_ePDs?t=720) #career #junior-dev ID:rw_001 ### [H2] GitHub contribution graphs mislead hiring decisions > [!quote] Dev Career Secrets from a GitHub Strategy Leader > "I don't judge a person's capability by how many green squares" Quality of contributions matters more than commit frequency for evaluation. [view highlight](https://youtu.be/I9zQos_ePDs?t=920) #hiring #github ID:rw_002 --- ### ๐Ÿ”– Atomic Split Pad (Zettel Candidates) Split level: H4 #### [Z1] Public learning documentation creates career opportunities - Applies to developers building professional networks - Referenced at timestamp 22:10 - Links: [[Community-Building]] [[Career-Development]] #### [Z2] AI usage requires strategic positioning in job applications - Relevant for developers incorporating AI tools in workflows - Referenced at timestamp 35:00 - Links: [[AI-Strategy]] [[Resume-Optimization]] --- # ๐Ÿ“„ Full Transcript *Auto-generated from YouTube captions ยท timestamps roughly every 30 seconds ยท [watch on YouTube](https://www.youtube.com/watch?v=I9zQos_ePDs)* **[00:00]** the talk with Franchesco series and I have of course a super guest. Not not a special guest but a super special guest. Usually when I have people here on the podcast first we do the podcast and then we we meet in person and then we take a selfie with with Andrea. We did something. So one second. Okay. And um in this case we did the opposite. So we met in person multiple times. I think that Andrea almost got um let's say **[00:30]** bored of me last year. But [laughter] we are here [gasps] today of course to talk about developer careers 2026. I can't believe that we are at the end of 20125 that in three two weeks and half will be straight into 2026. That's mind-blowing. Welcome Andrea. How are you? Please introduce yourself. >> Hello Francesco. I could never be bored of you. You're my favorite Italian. So that would be a lie. No. Hi. Hi. I'm Andrea Griffiths. I am a developer **[01:01]** advocate at GitHub and I did have the pleasure of seeing Franchesco three times last year. Was it three times? Yeah. >> Uh yeah. So it was great to meet you. I mean obviously like I've I knew of you. I followed you online like a lot of us. Uh but it was so nice every time y'all if you're watching this and you know what I'm talking about. Every time I meet someone in person that I known online, it's always kind of a little bit scary because I have been disappointed. **[01:32]** They say don't meet your heroes and some people are just not who they show they are online. But you are the real deal. So I've been veryful. >> Of course she's forced to do to to say this now because she is on my podcast. But I will keep an eye on this if she calls me out. You know I don't you [laughter] know I don't I can't I can't lie to our amazing friends here. No chance. >> Perfect. Perfect. Before we keep going. Hi Razan. Hey Yorkdev. How are you? How is it going? Yes of course we have. By **[02:03]** the way it's by chance we are wearing zers by chance but hair is way way better than me. Of course. So Andrea uh a little bit about yourself and then I want to really focus this episode for us for us developers because I think it has been a tough year for most of us layoffs. So I want to try to give say something that can help as many people as possible. Feel free to drop questions related to career um getting a job in tech. I have a list of questions. If we **[02:34]** have some very cool questions we can also pick them. Okay, Andrea, please tell us a bit more about yourself. >> Yeah, so I I get to work at GitHub where it's kind of a wild wild thing to even think that I I get to work there and I get to support all of you. Uh which is great. I'm I'm I'm in a role that I really enjoy. Uh but I didn't start my path to uh computer engineering in a traditional way. I'm a career changer. **[03:05]** her church careers later in life. So, it's an interesting it's been an interesting process. Um, but I would say now I'm not so new new, you know, like I've been doing this now for almost 10 years. Uh, but it's definitely interesting to get started. Um, especially now, I think if I was trying to make that career shift now, it would probably be a lot harder than it was for me back in 2016 because times are very different. like new developers, even a **[03:35]** senior developers are facing a completely new market. We're on the what is it the fourth industrial revolution at this point and times are moving fast. A lot of companies are making a lot of changes. A lot of companies love to do this right before the holidays. So, shout out to all of you out there. If you're being affected by recent layoffs, I'm so sorry. Like, please connect with me via LinkedIn or here. And if there is any opportunity that I see out there that I can share with you, I'll be happy to. So there's a lot of uncertainty in our market. One thing that I think is **[04:06]** not uncertain is that we need developers more than ever. So hopefully the conversation that we have today kind of helps people get a little less doom and gloomy because I think we're all kind of a bit exhausted of AI, exhausted of everything that's going to hell, of all the negativity. So hopefully we can walk out of here today feeling like, okay, I'm here. There's a place for me in this industry and I can be successful. I have a very non-traditional background and **[04:37]** I've been able to grow a career that I'm proud of. Keeping it authentic, not telling lies about who I like on the internet. That's for sure, Franchesco. I don't lie. My face won't allow me to. You'll be able to see right through me. [laughter] So, yeah, that's a little bit about myself. >> Perfect. Thank you so much. And yes, I think we are all all a bit exhausted of so much AI. Not because AI is so bad, but I remember that I for some reasons I brought a talk I think last year that was not about AI because I think it was **[05:09]** Rust and web development. So I I started saying like I'm sorry but my talk is not about AI and everybody laughed in the audience of course. So this made me feel in person like what is the real feeling of developers. I think a great plus of going to conference is that you know it's not Twitter so you have you know the real vibes the real feeling of what people feel but okay so >> for sure >> yes Andrea let's start with addressing you know the elephant in the room I want **[05:39]** to start with the super spicy question which is like is the role of junior developer like uh like doesn't exist anymore like what's happening with the junior developers and this is I think the most important thing that I want to discuss with you because uh I think the barrier now is higher. I would like to know from you what what do you think about this you know the role of the junior developer? Yeah. No, no way. If anything, there is there is a **[06:11]** need for more developers for sure. And hi friends. Um, contrary to what some people on the internet, especially on Twitter, will have you believe, no one is born a senior engineer. [laughter] Like every single person had to start at some point as a junior. So, I don't think the role is disappearing as much as it is going through a shape-shifting moment, right? Like it's changing. uh the way that we're creating software now has changed. So **[06:43]** that's obviously going to change that entry point for a lot of people, right? So like you remember your first scenes that you used to do when you the first from like hello world, right? Like the first lines of code where you were like, "Wow, I I made a machine do something. Look at that." Uh and it's like great, you you can write for a loop, right? But now there is a higher bar. Um AI, believe it or not, has raised the way that we have to think about what it means to be a junior developer. And so that old part of like, okay, yeah, **[07:14]** you're able to write code and you can sit down and hack it out and here's a page like it's a bit irrelevant now because there is a lot of systems in place that made that entry point sort of table stakes. So now we can everybody can go and b code an app basically pretty much we're using the the right tools. Uh the question is can you debug that app? Can you put proper security measures in place in that app so that you're not losing your entire database **[07:46]** of customers if you're, you know, like we've seen the horror stories play out on tech Twitter, on Reddit of people that vipcoded things and then they're like, "Oh no." Uh, or even catastrophic where it's like you lose information, you're supposed people's information or you're getting a $50,000 AWS bill because you post a key somewhere. So I think AI sort of raised the bar in what it means to be a junior developer and that we're forced now to go into the field with a lot more context and thinking a lot more about okay this is **[08:19]** not just the low entry point table stakes like I have to actually know why things work. I have to actually know how to debug them. And I have to think more as an architect and as a systems person like design systems are so important right now. Uh because again anyone can go in and hop in one of the many many many tools that are out there to like just create a skeleton app. But if you want to create something that truly scales and something that's going to support a business or you're going to go work for a business supporting their **[08:50]** applications like you need to actually know what you're doing. So it's morphing. I think is shapeshifted. Um AI can do a lot of things for us. It can create these beautiful scaffoldings. It can autocomplete, but again it's not going to make those important decisions that really matter at the end like the security, the architecture of the things that you're building. So I think for junior developers right now more than ever it's important that we sort of take more ownership uh of of software development as a whole. Um, it used to **[09:21]** be that when you came in as a junior developer in the company, you were very pigeon hole like okay, you're the front-end person, you're going to be focused on this part and now like you really need to have a broader look, a bigger understanding of everything. Um, a bigger understanding of what it means to the business really things that we often don't think about because we're like, no, my job is to hack it out and write code and that's what I do. But to be really successful now, you have to have sort of that eagle view and take all those things into consideration. So **[09:51]** I don't think it's it's going to be a career that is expiring as much as it's a career that is changed a lot and now you have to come into the field with a lot more skill and thinking more broadly than before. um you know like even internships like interns are doing things now that before I I came to GitHub the first time as an apprentice and like I definitely did not have uh even the opportunity to be as impactful as some of like our summer interns are doing things that are like going into production changing the business so it's **[10:21]** it's it's a exciting time in a way I think for juniors because you can be a lot more impactful but it's also a harder time because you do have to come more prepared like you can't the time for learning the table stakes things is is it's got to before you enter the market. [snorts] >> Yes, thank you. Thank you, Andrea. And I think that this visualizes so all the say most of the problems that are actually in the in the tech ecosystem. So AI made basically everything easier and everything harder. So we are at the **[10:54]** starting point again because this is how it works because it's an ecosystem. And so if everything if everybody can bible code an app then probably a person who knows how to code properly will still make a better app than you. We have sharper knives but the chefs still cooks better than everyone else because for them this is a it's a job. Absolutely. Yes. So I think we I think we found a solution Andrea which is we just use uh AI to have a whole green GitHub profile **[11:27]** and then we are done. We'll we just get hired right what do you think about this solution? No, Franchesco. No. And honestly, and I I say this from the bottom of my heart, like I I don't take that into consideration, and I I've I've been very lucky that I have some I I've been I've been able I have friendships that I cultivated throughout the years with both recruiters that are recruiters who are non-technical at all and they're just looking at like, okay, yeah, when they get thrown a profile of someone, **[11:57]** they look at certain things. And then some recruiters who actually have a technical background and and would know what they're looking at when they look at something like your contribution graph, right? And honestly, that is never told the entire story. I don't judge a person's capability by how many green squares. I mean, as a matter of fact, I might even judge you a little, but not in a good way. Like, if you show me like something that is [laughter] super like panon green all the way, be like, "Wow, when where is where is the rest of you?" Like, **[12:28]** if this is all there is like where is there more to the person? So I don't think it's never told a full story. It's told a story but it's still a really unreliable story because people gamify it too like you know they they commit I mean yeah and anyone can do something to commit so there is a a a square in your contribution graph. um I look at more like real signals like the meaningful contributions and the pattern contribution is not always 100% code and that's another thing that I think it's **[12:58]** also very misunderstood like yes it's great that I can see this brilliant project that you created and all this obviously you were busy the month of December because that's when all your green squares are happening um but then maybe you open source it maybe you didn't maybe you work in a team maybe you didn't I mean looking at things like issue discussions like the way that you're actually communicating the value of your project, how it's documented, what it is read of your project looks like, uh how your code like your commits, what what is that conversation **[13:30]** happening even when it's a solo project like how are you yourself doing engineering best practices when you're approaching a project even if it's just your own project. So I don't think it takes the whole story. I think I hate it that I love it because it's such a cool thing. And listen, I'm a fan of the contribution grab. I got the mug >> to show it to prove to prove it. I want it. But but I don't I don't think it tells the full story though. Honestly, I don't think I I don't think the gold standard is is is >> quantity is quality should be value. Um **[14:01]** it's the same thing like if you see people that are online posting a ton of content, making a lot of tweets, and what are they really saying? they're saying nothing. Um, so I I I don't judge that as a good um bar to say this is a better engineer than the other. Um, I think looking through the way that you're reviewing a PR gives me a gold mine of information. Are you giving thoughtful feedback? Is the interactions where like and again the way that you're documenting this is so important because **[14:33]** it lets me know like if you're able to explain the way that you're designing something, you'll be able to explain it to the team. you'll be able to work with the team. If you if you're a person that's actually building in public and you're contributing to open source projects, like you better believe it. I'm going to go back and dig through those and take a look and see how those interactions happen. Um especially when I'm I'm in a position where I can give an opinion of what I hire because that to me says more, you know, are you a part of a developer community? Like how how are you interacting there? How are you treating people there? Um so even as **[15:06]** like PRs and your contributions the merge PRs like still there is so many things that they don't fully show. Um but I think that there there is a room for us to look at the contribution graph as a part of a person but not the whole story. So um I definitely don't don't judge the quality of of someone's contributions just because it looks like they made the most. Uh it's uh it's a sad thing that you know it's the same story that and please don't don't add me **[15:37]** that people that are like oh and they love to post pictures of like it's 11 o'clock at night and everybody's still in the office. I'm like holy crap that tells me I never want to work for you. It doesn't mean I didn't work at 11 o'clock sometimes or on the weekends or you know like I do because I do I I do take a lot of pride in my work and I do put myself in positions where it's like ah why'd you do it? But but um that that shouldn't be something that's glorified, you know. Um and it doesn't tell the full picture of like the quality of person you are both in engineering and in life. So **[16:08]** >> yeah, >> exactly. I'm definitely sleeping at 11 p.m. But then of course I still can do a lot of stuff. I can wake up early, but it's not something that should be glorified. I'm always a huge s sustainer of sustainability and um because this is what will let you stay in if you want to stay in this career for 30 years, 40 years depending on your age but I want to stay here for even after retirement. So perfect. Thank you. Thank you so much. Thank you so much Andrea. Now let's make **[16:38]** a quick h quick game. You are a recruiter. You have 10 seconds to check one profile because you have to check a thousand profiles at 100 profile and you have to dedicate just 10 seconds to each of them because you need you know you need to filter a bit because you want to have a life right. So what would you do? I think this is this information can help you know to at least be sure that you are at least checking this. So what is the thing you will do you know to say **[17:10]** you know the yes no you know the X factor thing in 10 seconds. Okay. So, I land on your GitHub profile right now and uh first thing I look at uh not the contribution graph. Um your pin repos, right? Like the repositories that you pin. Uh what what are you choosing to show me? That's that tells me a lot about what's important to you, right? Whether it be your your own projects or projects that you forked and you're working on. I mean, I think that that says a lot about **[17:40]** what your priorities are. Um your bio, right? Uh even that short little sentence on the about me kind of tells me what sort of how you want to present yourself and that's going to that's going to tell me a lot about who you are as a developer. And then I do look at the recent activity but then again not the volume of the recent activity but the type of the recent activity. So are you opening issues? Are you reviewing PRs or are you just pushing commits to get these green squares to repose that **[18:12]** nobody's actually going to see? Um I I would look at those things first and then that's like sort of looking for the the the entry point positive things but then there are some like okay like I will move immediately I guess from the from the profile if there is nothing if there is no information. And I know a lot of people are very there's still if you're coming into the industry, you're new, you're just setting up your GitHub profile, like there is literally you can take 10 minutes of your time to write an about you, write your readme profile um **[18:44]** and just give a little bit more context on who you are because whether we like it or not, recruiters do look at this, right? So, uh, the readmi profiles, there's a lot of really amazing ones out you can use for reference, but also there is some brand of like really icky kind of templated clearly copy and pasted and not giving attribution to the people who actually created them. So, that would be a turnoff for me. I'll be like, h never mind. Uh, and then a ton of like fork repos that just show like, okay, you a hundred different things, **[19:14]** but nothing really stuck. like there were there was nothing that actually showed up on the recent activity. Um and then because I'm me and I again back to the I want you to come as a whole person like I would love to know a little bit more about you beyond the code. Um, so maybe in if your private life is private, but you know, like what you're passionate about coding and and yeah, you're super passionate about type languages. Like I love that, but also like throw me a bone. Tell me you like to ride a bike on the weekend or **[19:45]** something like those kind of things kind of I I find very interesting. So, um, an empty profile I probably forgive quicker than a busy for the sake of busy sake one, but take a time to like make a profile that, you know, at least tells me something about you at first glance. >> Perfect. And of course, does it mean that if you have a super polished GitHub profile that you are the best developer in the world? >> No. But you are you are present you this is how you present yourself in so in these difficult times. If there were you **[20:16]** know job for everyone, everyone had know time to make long interviews but this is not how it works and this is why it's so important. Think if you are here in the chat and you have coded a lot but you never p pushed on GitHub you are in trouble because it's the same thing of um playing guitar 10 10 hours a day but you never do anything. How are you people supposed to know that you are that good? That's the real That's the real issue. So, thanks. And I liked **[20:49]** especially the the pinned part that I think is absolutely underestimated. I think that if you're looking for a React role, you should have at least two GitHub repositories pinned that are about React because this is what >> exactly >> you should work because exactly this is where people check. It's the same reason why people check the pin tweet or what you post on YouTube on the first the first video. So the people check the pins and stuff because they don't have time maybe to check 100 repositories and uh yes and also and also I liked the **[21:20]** fact not too many things if you are an expert in everything you are an expert in nothing I remember my first CV was that long no probably I mentioned two things last JavaScript that's it because now yes exactly because now now you know exactly what you want to work on >> you can say I can do everything you be you your value will be perceived as as a as a lesser. Maybe you are an expert better than everyone else on on many technologies but still maybe you are **[21:52]** looking for something specific not just please give me any random job in tech. This is not how it works. Usually companies they look for a specific role react developer backend developer with a specific stack a specific library. This is why they'll give you money. This is why they give you a salary. So this is uh this is something like that I think it's important to mention. Thank you. >> Thank you. So maybe we can't uh we're never sure if we will get hired because this can also depend on the company they **[22:25]** needs maybe they listed they need a Python and JavaScript developers but in reality they need a Python developer. I don't I have no idea why this happens but this is the truth. So how can we make ourselves hierallable? So it's like okay like we can get we can get hired. So basically to increase our chances. >> That is excellent an excellent excellent question. Um [clears throat] I mean I think it that really sort of depends of what stage of your career you're in **[22:55]** right like where where you are. But if we're talking about for folks that are just new to the industry like they're trying to come in as junior devs. Um obviously the things that we mentioned earlier like having that good profile having well documented yes if you're applying for a job at Facebook to work on react I will hope that you either contributed or at least have one project that shows you you know like the skill is there. Um but also I think the the **[23:27]** the networking and the community aspect of it is sometimes sort of under under hype. Uh, I think it's it's underestimated how if you're in a more senior position in your career, nine times out of 10, you're not even like really applying for jobs. They're just you're getting sicked out like because a lot of these things come through networking, you know, like somebody like I know I know a really good Russ guy now. So if I ever have a Russ question or I have or I see a Rust related role, **[23:58]** I know exactly who I'm going to call. Um, so just kind of sort of picking the thing that you're known for and make sure that you're talking about it out there. And one thing I think uh we also underestimates ourselves is how many people are actually like reading and looking and watching the things that you're doing. Um, if you have a account on Twitter, forever Twitter, and you have 50 followers and you're passionate about talking, I don't know, whatever it is, like a remix, you think it's so cool. I want to talk about it all the **[24:29]** time. go there and talk about it. You never know who's actually looking at that and who's reading and where that opportunity might come from. Uh oh, but I don't have a 100,000 followers. It doesn't matter. Honestly, I've been so surprised at some some of the comments that I got from both people that I met in person and then online where they're like they'll mention something very specific that I did many years ago even and I'm like oh my like it blows my mind. I'm like how in the world did that even reach you? Um, so I would, you know, figure out the thing that you want to focus on and start putting yourself **[24:59]** out there as a person who who is clearly an expert in that field or is clearly learning in order to become an expert in that field. That's also another underestimated I think skill opportunity for you to storytell. Um, humans we love we love to follow a story. That's that's just the way it is. Like and I I love I'm I'm on the positive following the story, hyping you up. Like I want to see you succeed. there be haters out there that want to see people like not succeed, but regardless, we are glued to a story. So, folks that do a really good **[25:31]** job at documenting how they're learning, um, tend to be a lot more successful and and getting the exposure they need to actually at least get through this once you get through the scanners, get through that initial recruiter interview, then you have something to show. Uh, and being part of a community is a is a easy but also very rewarding way that you can sort of put your best foot forward there. So, um, I would say document everything that you're doing. I'm very, it's very hard for me to do that. That's something that I don't enjoy doing particularly, but I try to be more disciplined. And you have **[26:02]** something to say, say it. I don't care if 10 people saw your tweet or a million people saw your tweet. Um, you know, have opinions that are they're your opinions on them. And you know, sometimes they might be contrarian to what the coolest person with the most followers is saying, but if you feel confident in your skill and in your opinion, say it anyway. Say it anyway. Um, there might be a company out there that's looking for that very specific take on that thing. So, don't don't **[26:33]** don't follow the herd. Just own that own all of you. Um, that's one way that at least I think uh it would be it would be good for you. And then have your things online tidy like make sure you have your GitHub uh with a readme profile. I think Franchesco you need to do a how-to on that or or host maybe online that'll be a cool little like a display of good readies because some folks have done a really good job at exposing how great they are and you can just land on their GitHub page and you get like the full view. They're really really good at **[27:04]** that. My read me is I would say mid. It's not the best. I haven't put a ton of effort in it, but I feel like I got enough there that is like the things that matter to me uh that you see when you land on my page. So, yeah. >> Yeah. And I mean, it's absolutely fine. We should not get obsessed with our profiles 247. It depends. And when when we'll have, for example, now I have more free time. I'll probably check out the read me if it's up to date. But at least once in a while, you should check it. I love your portfolio. I love your personal websites website. **[27:35]** >> Oh, cool. I really love it. I like it's very simple, but it's very on it's really on point. I'm about to restructure part of my of my portfolio so I can get even inspired by it. I really like like it. We will leave the link below so you can check it >> and uh yes. Yes, >> absolutely. Yes. So now let's talk again about AI. AI and developers because I think developers they have this sort of like know love and hate with AI because they **[28:07]** want to use AI to improve their career even to apply I mean you can even use apply to >> reason but then you don't know where to stop when you should you know when should the human actually show so what do you think is the best use of AI for developers who want to get hired so basically to to improve their career, especially in in this process. What's the best way to use it? Let's just uh I don't know, let JGBT do everything and **[28:39]** you'll answer every email by clicking the AI button. >> No, [laughter] no, no. I think and and you you you've said it already like if you're using it as a tool for learning and growing, like amazing, for sure. Um I think there is a lot of misconceptions out there like it's not all bad and even actually most a few months ago I was taking a look I had um a fantastic open source project on a open source Friday stream. They're called just not just a J just another **[29:11]** jobseeking app. It's J AJA for short. I'm going to find the link and share it with you Franchesco so you can share because it's actually a phenomenal project. Um I'm sorry. Let me mute this dog real quick. >> It's fine. It's fine. >> I hope you're enjoying this uh this live event with Andrea. Andreas is really amazing. >> It's like you you just This is how >> Okay, you're muted now. You're muted now. I can't hear you. You're muted. **[29:47]** You're muted. Don't know why. >> So sorry. Give me two seconds. >> No problem. No problem at all. In the meanwhile, I can check the chat. So let me know if you are here. I have a question by was coder and conventional background medicine, economics boot camps. Amazing. You have background in medicine. No degree in computer science. You can absolutely find I think a um job in um in computer in um as developer **[30:18]** scholarship or opportunities. Uh I don't know about a specific one but for sure there are I would really push in something that is related to my domain because if you have a domain in medicine and economics you can I think you I would push in that direction. So I'll put in that I don't have a specific thing. Yeah. So sorry, Franchesco. There was somebody standing right in front of my window. So he was not going to stop. He was like, "Hello, mom. Someone is trying to break into our house right **[30:49]** now." Like somebody they see me. And I signal like I'm see I can come to the door and they're like tapping on the window. [laughter] >> So anyways, uh I'm sorry about that. Uh but yes, let's uh let's resume. I I I'm trying to think of my train of thought. We're talking about using AI. Yes. couple of months ago this project that I had online um for there they created this extension because they were looking for jobs and you know how you apply to like a million jobs and then you can keep track on your email of the applications you sent who reply who **[31:21]** actually like set up an interview with you but they created this open source project that does that and it's called just another app I'll find the link and share it uh but the reason I I thought about this it's um because while we were doing the the the show we were looking through like the different roles that are open out there and some of the roles are for companies that are basically AI AI companies like Antropic is it was hiring at the time for a ton of roles. They're still hiring for a ton of roles and I love how they break it down **[31:54]** what how could they use AI as part of your application process because they're being realistic about it. You're not going to not use AI, right? Like if I if I'm trying to tailor my resume to a particular role, like realistically, I'm going to probably use AI to help me at least dissect and and do it faster than I would. So, they have a really cool guide on their website uh when you're looking through their career page that actually tells you like this is what a good usage of AI is. Basically, it's not saying don't use it as a point of of of **[32:26]** a starting point or point of reference, but please don't create your resume entirely with AI because it is obvious. Like you can 100% tell when when something has been literally a copy and paste it from uh just chat GPT. Uh, you know, I've seen things where people forget to remove the like the either the footer or the top and then it's like it's painful because I bet you actually maybe sometimes you even worked a lot on that. Uh, but then when it was time to send that, you just hit copy and paste and there you go. Now I know that like I **[32:57]** feel like it was less effort effort wasted. So I think there is ways for you to use AI in a way that is for your advantage. I definitely uplevel my skills, uh, my technical skills on things that I was not super strong on with the use of AI. You know, having the conversations with the assistant more than just letting it write everything for me. Um, it's it's helped me level up in a lot of just things that were not my feel or a muscle that I wasn't flexing all the time. Like I don't do front-end design. Like thank you for complimenting the website. like my first iteration of **[33:28]** the website was [laughter] not great because I am not I'm not a front-end person. Um so I've been able to use AI to just sort of at least level up on those skills. That's one good usage. So if you're learning Rust and you're like using this uh coding assistance like use them more as as that use them more as uh leveraging for you to kind of be be a conversation and for you to fill up the gaps and skills and then when it's time to actually apply for a job yes you can use it to uh sort of validate your **[34:00]** resume your experience but adapt it but not as a copy and paste situation. And then if you're using AI, um I think we're missing that that's actually a good opportunity for you to put that in a resume in a way that is in a positive light like not as a negative. Um because I'm not going to I mean I would be hardpressed to look at a at a resume now this today in 2025 and if there's someone in the chat right now who's never used any kind of AI for coding **[34:30]** please let me know because I I love to meet this unicorn because I'm like no there's that like one way or another even if you don't know it there's been somewhere in automation there has been somewhere powered by AI that you've used so own that narrative right like frame it more as a it was a tool that I use to verify the judgment that I was making or help inform a decision or help me expand the view on something like I'm telling you right now these are things I'm not skilled at. So I am going to sort of use it as a rubber ducky and kind of help me **[35:02]** expand my knowledge there. And so if I need to list that on a resume, I'm going to list that on a resume, but I'm not going to write like I use GitHub Copilot and that's the bullet point on my resume. Um that's like saying like I use Google [laughter] everybody does it but it's more like okay well I use GitHub copilot uh and AI assistant development focus on the design of my page and I used it leveraging my architecture decisions that I had made for my page and this is the specifications that I **[35:33]** created and using AI assistant development I was able to I don't know cut the production time on X Y or that um or I created these validation workflows for this AI generated code in production. So there are ways that you can leverage the things that you're doing with AI in a positive light and a light that actually shows a a prospective employer that yes, you're a person that is you know your staff, you know your fundamentals, but you're also **[36:04]** open and eager to work with the technologies that I think every company right now has a mandate to use AI in one way or the other, especially if you're a public company cuz investors love to be with the buzzy trends and the you know that that's what's hype so that's what they want to hype. Um, so you know, like there's a way that you can position your usage of it in a positive light. Uh, and then also like use it as a as a tool for you to learn. Um, not to replace your fundamental learning, your fundamental practices. Like I've been poking around **[36:37]** trying to do some things with Rust for ever and ever. and and I've had I have some really smart people, shout out to Miguel Ras, who actually tried to help me. Uh, but not consistently. I I I was just like, you know what? I I just don't I don't I don't get it. Like, and I've done the same Robert Ducking thing that I would do with Copilot, trying to look at some skilled things on Rust, but I watched your boot camp and a lot more things clicked because it was like actual dedicated time for learning. So, **[37:07]** you know, yes, it's good to use it in in in a way that you're experimenting and kind of stretching yourself out there. Lean into that AI because it's always there. You know, that's that's when you have time and you can I can quickly have a conversation about like review this diagram and tell me how I could do this better or I do a lot of security things with AI too. like I but with params that I've collected with my own specifications of what I know to be good security practices then I feed that into it and say okay review this and tell me where there are holes what could I do **[37:38]** better how can I improve this and it it's just a way to leverageing it um not replacing of course that fact that I needed to have that CSK I needed to know those fundamentals in order to know what I'm asking same thing with learning Rust I'm like oh I finally got a lot more just by actually watching an educational thing, but now moving forward like I can actually apply those learnings into the way that I'm I'm I'm using the tools. So I I wouldn't shy away from from using it and experimenting and using it to grow. But please please don't forget the fundamentals are so so important. And I **[38:10]** think the the persons that are coming into the industry that have that solid understanding of fundamentals and also the flexibility and eagerness to use the tools are the ones that are going to succeed long term. Yeah, I've been saying fundamentals are fundamental I think for the last 20 years and the more the year they pass the more it becomes more important. By the way, I'm not I'm not crying tears of joy, but I have something in my eye. I'm sorry, Andrea, but I don't know. Maybe I **[38:41]** stared at my screen for too long. So, I'm not it's not it's not crying tears of joy for what you said. [laughter] Um, yes. Thank you. I would I would like to you to try to say something about the specific about GitHub because of course you work there as developer advocate. Is there anything like in GitHub that can help you? If you have some some some examples that maybe are not known by many people but like how you can uh you **[39:12]** know is there any GitHub feature? Maybe you have something that comes up not on top of your mind. You're muted >> for [snorts] you to get higher or like if you're looking for a job or you're looking to learn. >> Yes. Let's say maybe a feature that not many people know about GitHub that can help them in their career. Let's say whatever comes up in your mind if you have anything. >> Oh, I got you. Okay. Um, so for your career visibility for sure, I would say the same things the same **[39:42]** points that we touched earlier. have a you know it is important to have a profile read me. Um a lot of people either leave it blank or they just write something like hi I'm Franchesco and I like coding. Okay that doesn't tell me a lot about Franchesco. So definitely taking the time to kind of stand out there to show what you're currently working on. uh when you show what you're working on that's like your you're exciting like that it it comes through like I can I feel like if I feel your enthusiasm for whatever it is like I know you're **[40:14]** enthusiastic how Russ um so sort of bringing that same energy to your read me is something that is a underestimated thing. Um, and then making it super clear on what you're looking for. That's a good space for you to do that, right? Um, that's a place where you can literally write whatever you want. So, you're like, I'm seeking for a role that will help me, you know, or I have this amount of amazing skill. I know I can be impactful in this type of place. Uh, that's is your read me. **[40:44]** You can write whatever you want there. So uh anything that you have that is outwards like we use a lot of link trees and different linking and if you have a personal website that's great. Uh but also this is a space where you can link everything. So if if you've been to a podcast with Franchesco put it there. If you are speaking at a conference link the conference and also at the same time you also helping increase visibility for that conference. So use that profile readme space. It's a good piece of real **[41:14]** estate where you can literally do anything that you want. Uh, and again the first thing that people see when they land on your GitHub page. So I think that's a underrated underrated underused skill. One thing that maybe is not as mainstream and maybe recently because I've been doing a lot of work in this particular in this particular point. Uh, it's something that I've been looking at niche for learning for upleveling my own skill. But it's also kind of made me think about, huh, this is an interesting way to see how people **[41:45]** approach uh CI/CD pipelines. I look at the GitHub actions workflow tabs in your repositories. I do. I do. I do because I've been in that YAML I'm going to stop hating on YAML and start calling it YAML hell because I feel like I'm setting myself up for failure there. This is where we are. So what we got? [laughter] So we're I'm going to say from now on I love YAML and I've been doing a lot of things that have required me to look deeper into the way that folks are integrating this. And so I'm looking I'm actually looking through your action **[42:15]** stuff and it does kind of made me think about a different angle of looking at how you think about quality um how you think about optimization. It's interesting to me to to parse through that. So believe it or not, there are nosy people out there that are looking at things like that. [laughter] So you know those those are just two things that you can do right away in GitHub. Um there is a very little known place I hope to be known by more people actually like our work in the team before it was what it is now. You know **[42:47]** we have a discussions board in GitHub like literally the GitHub discussions that is for the entire company that started as a discourse instance many many years ago that's when I worked on it. uh but it's now integrated part of the platform. So if you are thinking hey you know what I want to go learn something new or I want to know share or I want to have a conversation about a GitHub feature um go have it in public go learn in public with with there's a lot of great people there. That's also a good place to poke uh product managers and engineering folks from GitHub **[43:17]** because they do spend time there and they do read the comments. Uh I know a lot of people love Reddit and all that but I'm like I'm not going somewhere else. My code is on GitHub. my discussions about my code and GitHub. Uh, and if if you're if you're looking for a place where you can actually help out people and show what you know, that's a good place for it. I can share the link of it as well. Um, so that's another also underrated place. I know people that got in jobs because they have like they apply for companies uh maybe about a specific part of I don't **[43:49]** know GitHub as a platform or maybe about a specific um maybe feature or and under like this if I'm looking for someone to come and optimize my CI/CD pipelines like I'm gonna look at your actions on your own on your own projects but also I might look I'm also O going to look at your discussions and see if you have actually if you have had these conversations in the open and I'll share I'm sharing the links with you now on **[44:19]** the private uh site that's the link for the discussions um and then I'll share the the link for the uh just a job app because it's a great project and everyone who's looking for a job should definitely leverage it um so yeah so it's it's uh it's I'm going to look through that um >> everything that you're doing in the open it tells me a lot more about how you're going to work in my team and that's maybe an underrated type of thing. Uh and honestly there is a lot of spaces **[44:49]** now for you to kind of voice opinions and I would I would recommend keeping them in there as well because there is it's one less place that your recruiter has to go find. [laughter] >> Perfect. Thank you. And uh so watch out be careful because everything you do online will be seen in some ways. This is not just GitHub but let's say for GitHub it's a the professional uh place to where you can also add you know a bit **[45:19]** of your personality of what what who you are. It's a mix between I like what you said like not just write I I I I write code. Of course, we are giving these tips if you're just getting started. Once you have on have been working in the industry for 20 years, probably you don't need to mention everything. We are trying to optimize these tips for who is just getting started. They are struggling for finding a job once the world knows you. I don't I don't think that Leno store needs to explain who he is. He doesn't need anymore. It's the **[45:49]** same thing when we are on social media. Once you have a million followers on YouTube, your Twitter header, nobody could could care less. But if you are just at the beginning, it might help you in getting a bit noticed in what you do. So I really like that your your approach because it is for for the be for the beginner. The internet is forever. You are right. And now I think everyone is checking their own GitHub profile. I can check it because now I'm live, but I will check it of course later because I **[46:19]** think that it's important to check it once once in a while. Andre, I don't know if when you have to go. I will just before you leave, I would like just to ask you what which technologies and frameworks I think you heard are the things to bet on in 2026. We are at the end of 2025. What do you think are the technologies that developers should bet on if we are betting our time as money? [snorts] >> Francisco, are you trying to get me in trouble here? Like no **[46:51]** >> technologies technologies are like babies like you can't you can't [laughter] I can't say I have a favorite one uh somebody in the chat say definitely not Russ [laughter] or >> I love it I love it okay like so we do this report every year at GitHub it's called the state of the octoverse report I can share a link to it as well um to you so that you can share on the chat but it's a really good report in that it gives you a sort of highle overview of the what we know and and if you think **[47:22]** about what GitHub knows like we have the power of data. Um so it tells you you know like the the growing technologies the what programming languages are like the most used in GitHub what projects uh or communities or even parts of the world sold the biggest growth. So there is a lot there. Um, I'm not saying go look at this report and use it as your like bible like everything that oh it says TypeScript just surpass Python so **[47:54]** that's it. I'm going to be full on TypeScript. No. Um, I think it does give you a good show of like the industry as a whole of what's growing, what's sort of what's what's what's changing. Um, I think right now we're seeing a lot of growth that's driven by frameworks that are doing things like scaffolding TypeScript by default because AI and AI assistant development that benefits from that. Uh, but there is also some interesting finds in that report like there is there was like a tiny tiniest **[48:26]** meion of a race on the uh usage of cobalt which is such an old it's like 60 is it 65 years old? It might be. Yeah, I think it might be 65 years old. Um, it's a very old very old programming language. Uh, but there is like there was a growth on the active developers that are actually using that technology because now they're with the help of AI we're looking at modernization of some of the things maybe things that weren't touched from years and years ago because that's the way that it work and the **[48:58]** people who were working on it alone retire. But now with the help of AI, we're actually being more realistic about what can be modernized. So it's kind of interesting. I think I wouldn't pin it so much on one particular language or one particular framework. Although I got to say I've been looking at Tanstack and it's so good, y'all. Like I was looking at a comparison. Somebody wrote a really good article of a comparison of that in Nex.js. Um and it's I I mean it's that team is cooking. **[49:28]** They're doing a lot of great things. I think a lot of it is going to be more I wouldn't focus so much on one particular programming language or one particular framework. Um if you want to look at one interesting space and you're thinking of like maybe I haven't decided what I want to do. Uh one thing that we can definitely distill from that October report is the growth of infrastructure as code practices. That's that's the way the the the world is spinning now. Uh so **[49:59]** that might be an interesting space for you to take a look at if you've if you never actually kind of dug into that before. So I wouldn't pin it so much onto like oh this is the upcoming language although hey shout out to all of you crop stations out there like the ROS community is also cooking. Uh but something like infrastructure or code as code uh systems programming if you want to be ahead of the tool may ahead of the the curve maybe looking more at the not so much the things that are like super shiny and new and and glamorous but more **[50:30]** the things that are actually in production and cooking like uh I so I can't say one particular language or framework I would say more look at the technologies as a whole and then the sleeper the sleeper out there looking at at the systems design spec and infrastructure as code practices because those are the things that are that have seen the most significant growth and that's not going to that's not going to change like that's only going to keep shooting upwards. >> Perfect. Thank you. Thank you Andrea. And something about unstuck might come **[51:02]** up soon on this channel but I'm not seeing a single more word on it. I'm not seeing anything of it, but let me cook and let me do my own stuff when I'm not recording uh podcast. Of course, going live. >> Uh Andrea, I don't know if you have to go if you have uh something coming up now or or not. >> I got I got like five minutes. I So, friends, for you who speak Spanish, please join me on the quack which will >> drop the drop the let me drop **[51:33]** >> drop the link. >> Drop the link. Ordev lives in Spain. So it might might actually join. [laughter] >> Yeah. Oh yeah, there you go. I mean you can practice your your Spanish and I don't know if you have a big Italian follow in Italian and Spanish is very similar. >> Well I there is this stereotype that Italians can understand Spanish and it's quite true. Like I never studied Spanish but I still can understand 75% of the Spanish. >> Same [snorts] same. >> Sort of. You like you sent me the link **[52:05]** here. Perfect. >> Okay. Also, I am I'm recording a podcast after this. I recorded a podcast before that before this live stream. So, we're always busy, but um I really really enjoyed this conversation, Andrea. And >> thank you, Franchesco. Thank you so much for having me. Like, I this was fun. >> Uh please, friends, if you're watching this now or after the fact, like reach out. like um my DMs are open for real **[52:36]** and if there is anything that I can help you with like I'm happy to try. Um and come come to the next come to all the things like we'll have this we do on Thursdays we do streams uh later on Cassidy will be live uh with the rubber duck Thursday stream. >> Uh I will be doing the rubber duck Thursday stream but in Spanish coming up next and then of course on Fridays is open source Friday every Friday. So come come support open source. Perfect. And thank you so much. Of course, they were inviting me, I think, two times. Yeah. **[53:08]** >> Yes. Yes. Please. Oh, and please, please read this thing that I'm writing so that I keep writing it >> because I I I wanted to do something that was >> um that could be fully opinionated and not have um to go through the layers of things that have to you your friends. I mean, I'm sure you know, you wouldn't believe the amount of layers of things have to go through before they're hit the out public when you work for a company. So, I'm I'm writing my own newsletter now and I'd love for all of you to give it a look and give me some feedback on it. That would be great. **[53:39]** >> Perfect. We'll leave the links and Andrea, thanks again for being here. Thank you so much for everyone who came here in this I think it has been a while since I haven't done like a live podcast but I've been quite busy with conferences but I really wanted to have Andrea before 2025 ended and I achieved my goal so Andrea we made it. Thank you. This made me very thank you so much. >> I appreciate you. Thank you so much everyone who joined. Um sorry I was I **[54:10]** wasn't able to like reply on the chat but I appreciate so much. It's fine. See you the next time. Bye-bye. >> Bye. >> Uh, where is the button? Here. No, here. It's already live. How do I stop it? Interesting. See you all. I think it's bugged. Let me check. How do I