Prepare for SheBuilds before the hackathon

Prepare for SheBuilds: before the hackathon ![rw-book-cover](https://readwise-assets.s3.amazonaws.com/static/images/article4.6bc1851654a0.png) 🔗 **Source:** [Prepare for SheBuilds: before the hackathon](https://www.youtube.com/watch?v=N8W1eI1B_-0) > [!note]+ Document Notes > tags: [hackathon-preparation, product-management, no-code-development, lovable-platform, audience-research] > author: Deafany (Stephanie) > source: YouTube — SheBuilds > reference: https://www.youtube.com/watch?v=N8W1eI1B_-0 > date created: {{date:YYYY-MM-DD}} {{time:HH:mm}} > > # 🏷️ Prepare for SheBuilds: Before the Hackathon >  ## 🧭 Table of Contents - [📝 Summary](#-summary) - [📖 Main Content](#-main-content) - [🗒️ Notes](#️-notes) - [📚 References](#-references) - [🗣️ Paraphrase](#️-paraphrase) - [📰 Newsletter](#-newsletter) - [✅ Action Items](#-action-items) - [🧪 Hypotheses](#-hypotheses) - [🎯 Conclusion](#-conclusion) - [🔄 Refactor Appendix](#-refactor-appendix) --- ## 📝 Summary - **What happened:** Deafany (Stephanie), a Season 2 SheBuilds alumna and product manager, delivered a pre-hackathon coaching session to Season 3 participants covering branding, audience research, dashboard building, animations, and API integrations using the Lovable no-code platform. - **Why it matters:** Participants have 48 hours to build a real product. Preparation done *before* the clock starts can be the difference between shipping a polished product and getting stuck on logos and login flows during the sprint. - **Key stakeholders:** SheBuilds Season 3 participants, program leads (Whitney, Elena, Xavier), Lovable platform team, alumni mentors. - **Risks and tradeoffs:** Spending precious hackathon hours on branding, API setup, or security issues that could have been resolved in advance; the temptation to over-engineer (gradient logos, undecided auth flows) instead of focusing on validated audience problems. - **Bottom line:** Do your brand, audience research, integration testing, and animation planning *before* the hackathon. Treat it like a real work project. Failure is a data point, not an identity. --- ## 📖 Main Content ### 🎯 Core Argument Deafany's central thesis is that hackathon success is front-loaded: the preparation window before the 48-hour sprint determines product quality more than raw coding hours. She frames this through her own Season 2 experience — she did not win, but she extracted enough lessons to advise the winners, grow as a PM, and launch a women's community in Prague. ### 🧠 Think From Your Future Winning Self She opens with a mental model: imagine you have already won. What would that future self tell you to do *today*? This reframes mundane prep tasks (create social links, set up a LinkedIn, generate brand assets) as strategic, not administrative. ### 🎨 Step 1 — Branding Before the Sprint - Choose three main brand colors and a font *now*, before the hackathon starts. - Load those exact values into Lovable via explicit prompts; the platform can copy colors/fonts inconsistently if not specified. - Use Canva to create a brand kit; do not improvise gradients live during the sprint (she lost 20 minutes to this). ### 👥 Step 2 — Audience Before Solution - Resist the impulse to start with a solution. Pick a community you are already passionate about. - Run even a minimal survey — three conversations can surface patterns. - Scrape competitor Discord channels, Reddit threads, and LinkedIn using tools like **Apify** (free, community-built scrapers shared on Reddit) to understand what audiences are already saying. - Her own product (teaching AI to kids) hit a wall because she underestimated the friction of parental consent and selling to minors — an audience insight she could have caught earlier. ### 📊 Step 3 — Build a Market Analysis Dashboard in Lovable - Before writing a PRD, build a standalone research dashboard in Lovable (she built hers in under 45 minutes). - Her example: she scraped Lovable's own Discord feedback channel and produced a feature-by-feature breakdown, a trust-breaking problems analysis, and a prioritization view — all inside Lovable. - This replaces the instinct to jump straight to PRD writing when you do not yet have deep market context. ### ✨ Step 4 — Plan Animations Early - Animations are underrated in hackathon judging; judges visit your website and visual delight matters. - Lovable does not add animations by default — you must prompt for them explicitly (hover effects, magnetic scroll, 3D card fly-outs). - Reference: the Season 2 winner's site "Get to Give" uses 3D card animations that feel subtle but are highly polished. ### 🔌 Step 5 — API and Integration Decisions Before Day One - Decide your authentication flow (Gmail login, magic link, waitlist, demo bypass) *before* the sprint starts. - **Brevo** (formerly Sendinblue) is Deafany's recommended free email integration; Lovable natively suggests it and walks you through obtaining the API key. - Two-step Brevo setup: (1) get API key from settings, (2) deactivate API key authorization in Authorized IPs so Lovable can connect without extra authentication overhead. - Lovable can generate and preview onboarding emails — including conditional variants based on whether the user provided social links. - Waitlist data is stored in Lovable's files; ask Lovable directly where to find it or how to export it. ### 🛡️ Security - Lovable flags security issues automatically with severity levels (medium, high, etc.) and offers to fix them. - Best practice: write security intent directly into your PRD prompt, e.g., "visitors should not be able to access data entered by other users." - Security issues tend to appear as you build — they cannot be fully pre-solved, but anticipating them reduces panic during the sprint. - SheBuilds is hosting dedicated security micro-webinars during the week. ### 💬 Q&A Highlights - **Should you share the research dashboard with judges?** No — the product itself should reflect your research. The dashboard is a tool, not a deliverable. - **Personal email for testing?** Fine if you have no privacy concerns; create a test account beforehand to avoid losing sprint time. - **Brevo + personal email:** Brevo can send from your personal Gmail address (email is routed through Brevo but the sender address is yours). - **Spend money during the hackathon?** No — use free tiers and demo modes throughout. - **Lovable Cloud for user data:** Lovable Cloud (backed by Supabase) can securely store user data and scale; you can also connect directly to Supabase if preferred. --- ## 🗒️ Notes #### ⏱️ [00:00:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=0) Whitney introduces Deafany - Frames the session as alumni giving forward. - Sets tone: failure is acceptable, learning is the point. #### ⏱️ [00:01:30](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=90) Deafany opens — Season 2 failure framing - Did not win Season 2; reached out to winners post-hackathon. - Framing: "this will not be your last product" — reusable career perspective. - Mentions building Women Who Build (Prague dinner community) as a direct outcome of SheBuilds inspiration. #### ⏱️ [00:03:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=180) Future-self mental model - "What would your future winning self tell you to do today?" - Specific examples: social links, LinkedIn profile, brand assets, AI-generated video content (Grok mentioned for video). #### ⏱️ [00:05:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=300) Step 1 — Branding - Three colors + one font in Canva before the sprint. - Explicit Lovable prompts for exact font/color reproduction. - Lost 20 minutes to gradient logo during Season 2 — cautionary tale. #### ⏱️ [00:08:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=480) Step 2 — Audience research - Pick community first, solution second. - Kids product failure: parental consent friction not anticipated. - Tools mentioned: Apify, Reddit scrapers, Discord scraping. - Coach insight ($1,000/month): survey people about their *day*, not your solution. - Gender observation: women tend to take others' requests seriously but go easy on themselves — apply work-project rigor to personal projects. #### ⏱️ [00:13:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=780) Step 3 — Market analysis dashboard in Lovable - Built in under 45 minutes using scraped Discord data. - Outputs: feature breakdown, trust-breaking problems, prioritization view. - Replaces premature PRD writing for participants who lack deep market context. #### ⏱️ [00:17:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=1020) Step 4 — Animations - Hover effects, magnetic scroll, 3D card animations. - Lovable can execute these — but only with explicit prompts. - Get to Give (Season 2 winner) cited as animation reference; team is in alumni Discord. #### ⏱️ [00:20:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=1200) Step 5 — API integrations and auth - Decide login method before the sprint: Gmail OAuth, magic link, waitlist, demo bypass. - Brevo: free, Lovable-recommended, sends from personal Gmail. - Two setup steps: API key + deactivate authorization requirement. - Lovable generates and previews conditional onboarding emails natively. #### ⏱️ [00:26:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=1560) Q&A — dashboard, emails, security, data storage - Research dashboard: tool for builder, not a judge submission. - Security: Lovable auto-flags issues; encode intent in PRD prompts. - Lovable Cloud / Supabase: viable for user data storage, scalable. - Waitlist data: stored in Lovable files; ask Lovable for location. #### ⏱️ [00:35:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=2100) Closing — mindset - "Everyone's a winner" — reframes the 48 hours as a learning artifact regardless of placement. - Rest when needed; give full effort in preparation. - Women Who Build community in Prague mentioned as post-hackathon outcome. --- ## 📚 References - [Lovable](https://lovable.dev) — no-code AI app builder used throughout the hackathon - [Brevo (formerly Sendinblue)](https://brevo.com) — free email integration recommended for waitlist/onboarding emails - [Apify](https://apify.com) — web scraping platform for Discord, Reddit, LinkedIn audience research - [Canva](https://canva.com) — brand kit creation (colors, fonts, assets) - [Supabase](https://supabase.com) — backend database underlying Lovable Cloud - [Get to Give](https://gettogive.com) — Season 2 SheBuilds winner, cited for animation quality - [Grok](https://grok.x.ai) — mentioned for AI video/content generation --- ## 🗣️ Paraphrase - Deafany's core message is that 48-hour hackathons are won or lost in the days before they start. Participants who arrive with a defined brand, a validated audience, a working API integration, and a clear animation plan are able to spend their sprint hours actually building instead of making foundational decisions under time pressure. - The secondary message is a reframe of failure. Not winning Season 2 was not the end of Deafany's product journey — it was the beginning of her growth as a PM, an independent community builder, and a mentor. The hackathon's real product is the participant, not the submission. --- ## 📰 Newsletter **🛠️ Build Smarter, Not Faster: Pre-Hackathon Prep Lessons from SheBuilds** If you've ever sat down to build something and lost an hour to picking the right shade of blue — this one's for you. Deafany, a Season 2 SheBuilds alumna and product manager, shared a refreshingly practical session with Season 3 participants: the best 48-hour hackathon starts days before the timer begins. Her checklist? Lock in your brand kit (three colors, one font — in Canva, before day one). Choose your *audience* before your solution. Build a market research dashboard in Lovable using scraped Discord or Reddit data from competitor communities. Decide your auth flow in advance. And don't forget animations — judges visit your website, and Lovable won't add hover effects unless you ask. Her tool stack for the prep phase: **Canva** for brand assets, **Apify** for audience scraping, **Brevo** for free email integrations, and **Lovable** for everything else (including a 45-minute market analysis dashboard she built live). The deeper lesson: treat your side project with the same rigor you'd apply to a work deliverable. And if you don't win? You've still shipped, learned, and — in Deafany's case — launched a women's dinner community in Prague. Not bad for a "failure." 🔗 [Watch the full session](https://www.youtube.com/watch?v=N8W1eI1B_-0) --- ## ✅ Action Items #### DevRel - Create a pre-hackathon prep checklist (brand, audience, integrations, animations) as a reusable resource for future SheBuilds cohorts. - Develop a short-form "market dashboard in Lovable" tutorial based on Deafany's 45-minute workflow. - Publish a Brevo + Lovable integration quickstart guide with the two-step API key setup documented. #### Community - **Contributors:** Pin Deafany's session recap and key tool links (Apify, Brevo, Canva, Get to Give) in the Season 3 Discord. - **Maintainers:** Invite Get to Give winners to host an animation-focused micro-session; they are already flagged as alumni in Discord. - **Sponsors:** Document the "Women Who Build" offline community model (Prague dinners) as a potential replicable chapter format for SheBuilds expansion. #### WordPress - **Training:** The audience-research-before-solution framework (survey → scrape → pattern → PRD) is transferable to WordPress plugin development and theme design; consider a lesson plan adaptation. - **Community:** The women-in-tech focus and "give forward" alumni model maps well to WordCamp mentorship tracks. #### The WPCC - **Programs:** Deafany's session demonstrates high value from structured alumni-to-cohort mentorship. Consider a formalized "alumni mentor" track with a defined session format and follow-up loop. --- ## 🧪 Hypotheses 1. **Preparation quality predicts sprint output quality:** Participants who complete brand, audience research, and integration setup before the hackathon will produce higher-rated products than those who begin cold on day one. 2. **Audience-first framing reduces pivot friction:** Teams that define their target community before their solution will require fewer mid-sprint pivots than solution-first teams. 3. **Explicit animation prompting improves judge scores:** Products that include intentional hover/scroll animations (explicitly prompted into Lovable) will receive higher design scores from judges than default Lovable outputs. 4. **Research dashboards accelerate PRD writing:** Participants who build a Lovable market analysis dashboard pre-sprint will write higher-quality PRDs in less time during the sprint. 5. **Auth flow pre-commitment reduces lost hours:** Participants who decide on their authentication approach (Gmail OAuth vs. magic link vs. waitlist) before day one will lose fewer sprint hours to login implementation indecision. 6. **Alumni failure narratives increase participant resilience:** Cohorts that hear structured failure-and-learning narratives from alumni will report higher psychological safety and willingness to pivot during the sprint. ### 📊 Methodology - **H1:** Retrospective survey of Season 3 participants comparing prep hours logged against judge scores and self-reported product completeness. - **H2:** Coded interview analysis of teams that pivoted mid-sprint vs. those that did not, correlating with pre-sprint audience definition documentation. - **H3:** Blind judging rubric analysis scoring animation presence against overall design score across all submitted products. - **H4:** Compare PRD word count, specificity, and time-to-first-build between dashboard-builders and non-dashboard-builders. - **H5:** Sprint time log analysis: minutes spent on auth implementation, correlated with pre-sprint auth decision documentation. - **H6:** Pre/post psychological safety survey (adapted Edmondson scale) across cohorts that received vs. did not receive alumni failure narratives. ### ❓ Questions - Does Lovable's security flagging system cover OWASP Top 10 categories, or only platform-specific vulnerabilities? - At what audience size does Lovable Cloud / Supabase require a paid tier, and what is the ceiling for free hackathon use? - Are Apify's community-built scrapers for Discord and LinkedIn within those platforms' terms of service for hackathon/research use? - What specific judging rubric does SheBuilds use, and how much weight is given to design/animation vs. problem-solution fit? - How does "Women Who Build" (Prague) plan to scale — and is there an open-source playbook for starting a chapter? ### 🔮 Future Research - Longitudinal study: do SheBuilds participants who do not place still go on to ship follow-on products at higher rates than comparable no-code learners? - Comparative analysis of winning SheBuilds submissions: what structural features (auth, animations, data layer, audience specificity) correlate with placement? - Survey of PM practitioners: how widely is the "build a research dashboard before the PRD" practice used outside hackathon contexts? - Effectiveness of scraping-based audience research (Apify/Discord) vs. direct user interviews for early-stage product discovery. ### 💡 Implications - **Practice:** The pre-hackathon checklist model could be standardized across accelerator programs and no-code bootcamps as a replicable onboarding artifact. - **Education:** The "future self" mental model and the "work-project rigor" reframe are teachable heuristics for reducing imposter syndrome in non-traditional tech entrants. - **Security:** Encoding security intent into PRD prompts (rather than patching post-build) suggests a prompt engineering best practice for no-code security hygiene. - **Funding models:** If alumni-led mentorship is this high quality, it represents a zero-cost program asset worth structuring formally (mentorship stipends, session recordings, reusable curricula). --- ### 🧠 Learning Styles Assessment - **Reading level estimate:** Flesch-Kincaid Grade Level ≈ 8–9. Conversational, accessible, light on jargon. Appropriate for a broad adult audience with no PM or technical background required. #### 👀 Visual learners - **Present:** Deafany references her Lovable dashboard and animation examples verbally; a website (Get to Give) is cited. - **Improve:** Screenshots of the market analysis dashboard, Brevo API settings screen, Canva brand kit template, and Get to Give animation examples would significantly increase comprehension. #### 🔊 Aural learners - **Present:** Full video with spoken delivery; conversational tone aids listening. - **Improve:** A 3–5 minute audio summary of the five preparation steps would serve podcast-style learners who want a refresher without re-watching. #### 🗣️ Verbal learners - **Present:** Rich in analogies (iceberg, future self, work-project framing), storytelling, and concrete examples. - **Improve:** A glossary of terms (Brevo, Apify, Supabase, magic link, PRD) for participants new to the stack would reduce friction. #### 🧩 Kinesthetic learners - **Present:** Step-by-step instructions for Brevo setup; live demo references. - **Improve:** A downloadable pre-hackathon checklist with checkbox tasks (create Canva brand kit, get Brevo API key, run Apify scrape) would be immediately actionable. #### 🧠 Logical learners - **Present:** Clear five-step structure with causal reasoning (if you do X before the sprint, Y will not cost you time during it). - **Improve:** A decision tree for auth flow selection (Gmail OAuth vs. magic link vs. waitlist vs. demo) and a risk/effort matrix for the five prep steps would sharpen the logical scaffolding. #### 🧑‍🤝‍🧑 Social learners - **Present:** Live Q&A, Discord community references, alumni chat callout. - **Improve:** Structured peer pairing (e.g., "find an accountability partner and share your brand kit by Friday") would create a social accountability layer. #### 🧘‍♀️ Solitary learners - **Present:** The "future self" exercise is inherently self-reflective. - **Improve:** A set of journaling prompts (Who is my audience? What would my winning self have done differently? What is my auth strategy?) would scaffold individual reflection. #### 🌿 Naturalistic learners - **Present:** Not explicitly addressed. - **Improve:** Framing the product lifecycle (idea → build → pivot → product 2) as an ecosystem / growth cycle could resonate with naturalistic thinkers. --- ### 📊 Accessibility Assessment #### Screen reader semantics - **Present:** Video format; YouTube auto-captions available. - **Improve:** Captions should be human-reviewed (auto-captions misrender "Brevo" and "Apify"); descriptive alt text for any slides or screenshots shared. #### Keyboard navigation - **Present:** YouTube player is keyboard navigable. - **Improve:** Any linked companion resources (checklists, dashboards) should be tested for keyboard reachability. #### Contrast and readability - **Present:** Not assessable from transcript alone; slides were referenced but not visible in transcript. - **Improve:** Ensure any published slide deck or resource page meets 4.5:1 contrast minimum. #### Text resizing and reflow - **Present:** N/A for video. - **Improve:** Any companion web resource should be tested at 200% zoom. #### Media and captions - **Present:** YouTube auto-captions exist. - **Improve:** A downloadable transcript (cleaned) and audio description of any screen-shared dashboards or animations would serve deaf and low-vision participants. #### Forms and interactions - **Present:** N/A for the video itself. - **Improve:** Any Lovable-built products participants create should follow the accessible forms guidance Deafany alludes to (error messages, label clarity). #### Standards reference - Gaps map to WCAG 2.2 AA: 1.2.2 (captions), 1.2.5 (audio description), 1.4.3 (contrast). --- ### 🌍 DEIB Assessment #### Cultural breadth - **Present:** Prague-based offline community mentioned; implicit global reach via Discord. - **Improve:** Explicit acknowledgment that hackathon participation spans time zones and that prep steps should be adaptable to participants with limited internet access (e.g., lighter-weight alternatives to Apify scraping). #### Inclusive language - **Present:** Gender-neutral for most content; "women" used deliberately in the program context, which is appropriate and intentional. - **Improve:** The observation that "women tend to go easy on themselves" is a generalization that may not resonate universally; framing as "many of us" or citing the observation as a pattern rather than a rule would be more inclusive. #### Neurodiversity support - **Present:** Clear five-step structure aids sequential processing; numbered steps reduce cognitive load. - **Improve:** Explicit "TL;DR" callouts per step and consistent heading use in any written companion would support ADHD and dyslexic participants. #### Representation - **Present:** Female PM and community builder in a technical role is meaningful representation. - **Improve:** The session could explicitly acknowledge participants from non-PM, non-tech backgrounds (e.g., teachers, caregivers, artists) who are equally valid builders. #### Psychological safety - **Present:** Strong — the failure framing, "everyone's a winner" close, and explicit acknowledgment of mothers building while caregiving create a psychologically safe container. - **Improve:** A clear code of conduct reference for the Discord and a moderated Q&A structure would formalize safety beyond tone. --- ### ✅ Logical Validity #### Strong reasoning - **Future-self heuristic (throughout):** A well-established implementation intention strategy in behavioral psychology. Framing prep tasks as actions your successful future self would have taken is a sound motivational device. Strong because it is directional and specific. - **Audience-before-solution argument (Step 2):** Deductive and evidence-backed by her own product failure (kids/parental consent). The causal chain (undefined audience → unexpected friction → failed product) is coherent and falsifiable. - **Pre-sprint integration decisions (Step 5):** Inductive reasoning from observed pattern (auth indecision costs sprint time) generalized as a rule. Sound, with clear actionable output. #### Fallacies - **Hasty generalization — gender observation:** "We as women tend to go easy on ourselves but take what other people are asking for us so seriously." Location: audience research section. This is a broad generalization from personal experience applied to all women; it may alienate participants who do not identify with this pattern. Fix: hedge with "many of us find..." or cite a source. - **Anecdotal evidence — future-product claim:** "This will not be the last product you build." Deafany presents her own trajectory as the expected norm. Location: opening remarks. Fix: frame as a possibility ("for many of you, this will be the first of many products") rather than a certainty. #### Tone of voice - **Warm, encouraging, peer-to-peer.** Examples: "I'm absolutely cheering for everyone," "whatever you build — you know — I think we don't give enough awards." The tone reduces hierarchy between mentor and participant, which is appropriate for the psychological safety goal of the session. --- ### 🛡️ DARVO Analysis - **Deny:** Not applicable. This session contains no defensive behavior; Deafany proactively acknowledges her own failure. - **Attack:** Not applicable. No adversarial dynamics present. - **Reverse victim/offender:** Not applicable. - **Note:** DARVO is not a relevant frame for this content. The session is a collaborative, alumni-led coaching context with high psychological safety and no contested claims. The failure narrative is voluntarily disclosed and constructively framed. --- ## 🎯 Conclusion Deafany's pre-hackathon session delivers a compact, high-signal preparation framework: brand before day one, audience before solution, research dashboard before PRD, explicit animation prompts, and auth decisions locked in advance. The meta-lesson is methodological rigor applied to personal projects — the same standards used in professional PM work. **Practical next steps for participants:** 1. Open Canva today; define three colors and one font. 2. Identify three people in your target community and send a survey this week. 3. Scrape a competitor Discord or Reddit thread with Apify; paste results into Lovable and generate a market analysis dashboard. 4. Open a Brevo account, get your API key, deactivate authorization, and test the integration before Sunday. 5. Prompt Lovable for at least one hover animation and one scroll effect before the sprint ends. 6. Write your security intent into your first PRD prompt ("visitors should not see other visitors' data"). --- ## 🔄 Refactor Appendix — Prepare for SheBuilds: Before the Hackathon ### 📚 Literature Atoms #### [H1] Pre-sprint brand decisions prevent wasted hackathon hours > [!quote] ID:rw_shebuilds_001 > "I was working on the gradient for like literally 20 minutes to get it right. These things could have already been done definitely." **Context & use:** - Timestamp ~00:05:30; branding section - Specific, recoverable cost of not pre-committing to brand assets. Applies directly to any time-boxed build context — not just hackathons. **Tags:** #xref #aphorisms **Highlight:** [view highlight](readwise://shebuilds_001)  --- #### [H2] Audience definition before solution prevents unrecoverable product pivots > [!quote] ID:rw_shebuilds_002 > "My product was for kids — teaching AI to kids — and later on I noticed how hard it is to sell something to kids or even get them to give consent." **Context & use:** - Timestamp ~00:09:00; audience research section - Concrete failure case: audience-specific friction (parental consent) only visible after build. The pattern generalizes to any regulated or intermediated user group. **Tags:** #xref #articleseed **Highlight:** [view highlight](readwise://shebuilds_002)  --- #### [H3] Lovable can generate a market analysis dashboard from scraped community data in under an hour > [!quote] ID:rw_shebuilds_003 > "I didn't spend more than 45 minutes on this and it created me a feature by feature breakdown of what Lovable needs today." **Context & use:** - Timestamp ~00:13:30; dashboard section - Establishes a concrete time benchmark and output type (feature breakdown, trust-breaking problems, prioritization view) for no-code PM research workflows. **Tags:** #xref #addendum **Highlight:** [view highlight](readwise://shebuilds_003)  --- ### 🧠 Permanent Candidates #### [P1] Encode security intent in the initial prompt, not as a post-build patch **Claim** (one sentence, in my voice): Security requirements specified in a no-code PRD prompt produce fewer vulnerabilities than security treated as a cleanup step after the build is complete. **Evidence pointer:** [[#H1 Pre-sprint brand decisions prevent wasted hackathon hours]] — security intent follows the same logic as brand decisions: pre-commitment reduces remediation cost. **Scope & conditions:** - Holds when: the no-code platform (e.g., Lovable) has an active security flagging layer and interprets natural-language security constraints. - Breaks when: the platform does not parse intent-level security prompts, or when the threat model requires formal security review beyond prompt-level controls. **Links:** - [[No-Code Security Patterns]] — because extends - [[PRD Writing Best Practices]] — because applies-in product management contexts - [[Lovable Platform Capabilities]] — because evidence-for  --- ### ✏️ Fleeting Queue - [F1] Investigate Apify's terms of service for scraping Discord and LinkedIn in commercial/educational contexts — is it within ToS for hackathon audience research? — https://apify.com/terms #fleeting #inbox - [F2] "Women Who Build" Prague dinner model — could this be templated as a chapter playbook for other SheBuilds alumni cities? — https://www.youtube.com/watch?v=N8W1eI1B_-0 #fleeting #inbox #storystem ### → Things 3 (max 2 actions) - [ ] Create a Canva brand kit template pre-populated with three-color + one-font slots for hackathon onboarding — share in SheBuilds Discord #things - [ ] Draft a Brevo + Lovable integration quickstart (two-step API key setup) as a standalone resource for Season 3 participants #things --- ``` Processed: Prepare for SheBuilds — Before the Hackathon - Literature atoms: 3 / 5 cap - Permanent candidates: 1 / 3 cap - Fleeting items: 2 - Things 3 actions: 2 / 2 cap - Konik tags used: #xref, #aphorisms, #articleseed, #addendum - Triage callouts populated: Cross Reference, Article Ideas, Vocabulary (empty) - Quality gates: ✓ verbatim quotes present / ✓ atomicity markers present / ✓ link annotations with relationship types / ✓ no [MISSING] strings / ✓ tagged highlights routed to callouts / ✓ view highlight links present - Override conditions triggered: none ``` --- ## Full Transcript [00:00] far. Um, and it's only the beginning. [00:03] It's uh it's like a an iceberg, right? [00:05] We're right at the tip and we're going [00:07] to keep on going. And it's all thanks to [00:09] our amazing alumni. Uh, one of whom is [00:13] here, of course, Stephanie from uh [00:16] season 2. And so, I'm going to shut up [00:19] and let her do her thing because I know [00:22] she's going to bestow upon you all so [00:24] much more wisdom. Um, and so please, you [00:28] know, join me in thanking you, uh, [00:30] Deafany for taking the time, wanting to [00:33] share this, to give forward to this next [00:35] season. Um, we really, really appreciate [00:38] your time. [00:40] >> Yeah, sounds great. Thank you, Whitney. [00:42] I'm super excited to be here with all of [00:44] you. So, hi everyone. Um, it was such a [00:47] thrill to join and be a part of season [00:50] 2. I remember getting accepted and it [00:53] was so exciting to meet everyone to [00:57] share all the ideas. So, I'm sure you're [01:00] having a thrill and you will learn so [01:03] much in this journey. So, I'm sure that [01:05] Elena and Whitney and Xavier have [01:08] already said this to you multiple times. [01:10] But definitely look at this as a opt [01:13] opportunity to learn and whatever [01:16] happens even if you fail that's the part [01:19] of the journey without failure there [01:20] won't be product number two product [01:22] number three and I can reassure you that [01:26] for sure this will not be the last [01:29] product that you build it will not be [01:31] your last pivot you will pivot multiple [01:34] times after this and build multiples of [01:36] products with lovable uh at least that's [01:39] what happened to [01:41] And I've saw that some of your messages, [01:43] some of your building communities, etc. [01:45] These are great way to start [01:47] understanding your audience and then [01:50] turn it into a product. So I think you [01:53] all are ready for it and I'm so excited [01:55] to see what comes out of it. And today [01:58] I'm going to share with you what I have [02:00] done as a failure story. So I didn't win [02:03] season two, but I learned so much from [02:06] it. I in the end reach out to the [02:08] winners as well to ask them what did [02:10] they do differently. So I'm going to [02:11] share a little bit of those as well. I'm [02:14] going to share a little bit of what I [02:16] think I do differently as a product [02:18] manager now today. You know looking at [02:21] back at my work. I have grown so much [02:23] since then in my in my job and also [02:26] externally. um you know doing some [02:30] workshops, speaking etc and engaging [02:32] with the community of women has been [02:35] absolutely my passion since then. So [02:38] let's get started and I'll share um sort [02:40] of a few steps that you can take before [02:43] the hackathon actually starts and really [02:45] practical steps. Um some of them are [02:48] going to be technical so maybe we can [02:51] even look at that how to integrate [02:53] things as well. [02:55] So um first part is again as I said you [03:02] should be thinking about your self as if [03:06] you have won already and what would this [03:09] self do best and what would they come [03:12] back from the end of season 3 let's say [03:15] you won and you delivered this amazing [03:18] project amazing product what would [03:21] yourself future self tell you today to [03:25] hey get this done today. Think about [03:27] those. Is it um creating social media [03:30] links already? Is it creating your [03:32] LinkedIn account so that you can post [03:34] later on? That also creates a little bit [03:36] of validity. You can then put those [03:38] links into your product as well. Maybe [03:41] it's creating the whole brand assets, [03:43] brand line. Maybe generating some videos [03:46] using AI like Grock is really great at [03:50] videos and video content. Um like one of [03:53] my personal learnings has been always [03:58] looking back and saying like these [04:00] little details, these things I could [04:02] have done earlier and my screen share [04:06] has stopped but we are still here. No [04:09] worries. And so basically [04:13] um I'm going to give you these little [04:14] technical tips about um what I do [04:18] differently and I think you should take [04:20] a look at. So number one is branding. So [04:24] we're going to talk a little bit about [04:25] branding. Um how to think about your [04:28] audience, how you can use lovable to [04:31] create a dashboard about your whole [04:33] market analysis, and then looking a [04:36] little bit more into the details. [04:38] So I think one of the things I really [04:41] lived through was, oh, okay, I don't [04:45] have my logo exactly. Let me put a [04:47] gradient into the logo because I think [04:49] that's going to look cooler. And then [04:52] working on the gradient for like [04:54] literally 20 minutes to get it right. [04:57] These things could have already been [04:59] done definitely. Um so think about the [05:02] colors. Um I was not really thinking in [05:06] detail. I think I could do much better. [05:08] Um I was just thinking like oh this will [05:11] go well you know and oh I like this [05:14] other website. I think this will this [05:16] lets me like just copy paste that, [05:18] screenshot that because that's what I [05:20] knew. But I think going into Canva and [05:23] creating a nice little model that really [05:26] is aesthetic for you, having three main [05:29] colors, this is going to really help you [05:32] uh grow the brand. Have a font already [05:35] and uh ask Lovable to exactly put that [05:37] font in because it can sometimes copy [05:40] differently. So make sure like it is the [05:42] exact font. ask it to double check the [05:44] color um etc. So have that brand exist [05:47] before you know the week already starts. [05:52] Um step number two I think picking the [05:56] uh the audience before you uh go to the [05:59] solution is a really important one [06:01] because we often try to start with the [06:04] solution. And for me what was really key [06:07] was going back later after the hackathon [06:10] and being like um cuz my product was for [06:13] kids. It was for um teaching AI to kids [06:17] and later on I noticed that how hard it [06:20] is to sell something to kids and or even [06:24] get them to you know um give consent [06:27] about a lot of things. You need to [06:29] involve parents. So really picking your [06:32] community that you are passionate about [06:34] already today. I think you're allowed to [06:37] pivot slightly. So you can maybe already [06:40] think about that if you can um or if [06:44] this is the thing you want to go for and [06:46] making sure you listen to the idea and [06:49] listen to the audience and where they [06:52] are and finding out what they are [06:55] thinking. You know what are their [06:57] problems? what do they need exactly? And [07:01] that's going to help you refine your [07:02] even if you have the idea and the [07:04] solution, that's going to really be the [07:07] key. Um, so for example, I work with a [07:11] coach for $1,000 a month for building a [07:15] business. And what she told me [07:18] essentially was and where I got blocked [07:20] before was really making a survey asking [07:24] people what they want, what are their [07:26] problems, what how can they resolve them [07:29] or how do they imagine resolving them or [07:31] even not even going to the solution. How [07:33] do they go about their day about you [07:36] know your product about the the industry [07:39] that you are thinking about? And this is [07:42] going to give you so many ideas and so [07:44] much patterns that going to come out if [07:46] you talk to just three people even. And [07:50] um when you once you find where they [07:52] hang out, you can also use some third [07:55] party tools like Appify or there are [07:58] some third party tools built by you know [08:01] engineers all around the world shared [08:02] across Reddit for free to use to scrape [08:06] um social media, scrape LinkedIn, scrape [08:09] uh Discord channels of products of [08:12] competitors to understand what the [08:15] audiences are talking about. So, I would [08:18] go ahead and do that already today. [08:21] Really see what people are talking [08:23] about. Do it as if you're doing it for [08:25] your work because I think often times we [08:28] also as women, we tend to go easy on [08:31] ourselves, but we take what other people [08:35] are asking for us so seriously because [08:37] we have this compassion to serve and you [08:40] know this intrinsic empathy to help [08:42] others. So, it can be really hard to [08:45] help ourselves. So, I really ask you to [08:49] take this project as seriously as you [08:51] would a work project and really hone [08:53] down on how would I go about this at my [08:57] work at my job. [09:00] And that brings me to my next point. I [09:04] think um what I have done for a lot of [09:07] other projects for work for interviews [09:10] in product management in tech u build a [09:14] standalone dashboard. This is going to [09:16] be your um let's say resource and you [09:20] can look at that uh and I will show you [09:22] one that I have done and just as I said [09:26] I have scraped uh lovable and lovable uh [09:29] feedback channel actually for the [09:31] lovable product and um lovable helped me [09:35] create um I didn't spend more than 45 [09:38] minutes on this and uh it created me a [09:41] feature by feature breakdown of what [09:43] lovable needs today. Um there's a deep [09:46] analysis into trustbreaking problems of [09:48] the audience that are writing actively [09:52] onto discord channel about what's [09:53] missing about lovable and it also gave [09:56] me a prioritization dashboard beautiful [09:59] like what's missing what do the [10:01] community say and this is so useful so I [10:04] think um if you you know if I were a [10:07] product manager this would be absolutely [10:11] amazing to decide what feature I'm [10:13] building next and I I think also as [10:16] product managers we tend to just [10:17] directly go to PRD writing or we hear or [10:20] if we're in tech as well we just imagine [10:23] okay I'm just going to write the PRD [10:25] there is a lot of strategy that goes [10:27] into the work that happens before [10:30] writing a PRD because if you're a [10:32] feature PM often you're aware of the [10:35] market you're aware of the customer [10:38] issues you're in it you know you're in [10:40] the thick of it so you know most of the [10:43] competitors what they're doing like [10:45] yesterday what they've done. So you can [10:47] write the PRD tomorrow but in this case [10:50] you are just starting this out [10:53] probably and you need all the [10:56] information all the details you can have [11:00] and lovable is actually offering this to [11:02] you already. So definitely pull this [11:04] out, create something like this and then [11:07] later on I even recommend that you [11:09] create um one [11:12] example of a prototype potentially and [11:16] see how you know lovable works. See how [11:19] you can um change different things, how [11:22] you can add integrations. So try things [11:25] out that you plan to use later on. And [11:28] that brings me to my next [11:31] um next point. [11:34] Last but not least, think about the [11:38] animations and the cool tricks that [11:40] you're going to have that are going to [11:42] be attention grabbing for well, not only [11:46] the people and your audience, but this [11:49] is a hackathon. The jury and the judges [11:51] are going to come into your website. So, [11:54] how can you make it really cool? And I [11:56] think one of the very um underrated [11:59] things is animations. You can animate at [12:02] a hover. You know, it can pop up. Like [12:05] my slide doesn't do this. But you know, [12:07] elsewhere, for example, [12:10] this little website that I've built for [12:12] my offline community here in Prague, we [12:15] do some dinner and drinks with women. [12:17] And you see, I did some little um [12:20] animations at Hover. And these Lovable [12:23] doesn't do by default. So, I'm just kind [12:25] of putting this out there that you might [12:26] be like, "Oh, I already know this." Um, [12:29] just think about what kind of animations [12:31] are going to be useful to you. It might [12:32] be a magnetic scroll that once you [12:34] scroll, it locks in. It might be, and [12:37] here's a really great example from the [12:39] winner of last year, last season. Maybe [12:42] you're aware of this website. It's [12:44] called Get to Give and it's so [12:46] beautiful. And they have really cool [12:48] animations. like you don't even notice [12:51] that when it happens, but these cards [12:53] come out and everything's kind of 3D [12:57] moving up and down. So, you can [12:59] definitely ask Lovable to do these [13:01] things, but it's just not going to give [13:02] it to you for free. So, just a little [13:05] reminder, think about these things. And [13:07] then my last point is going to be let's [13:10] take a look at how we integrate with API [13:13] keys and APIs and lovable because this [13:16] is also something that I played a lot [13:18] like back and forth in my head. Oh, am I [13:21] going to have a Gmail login? Am I going [13:23] to have a like a a magic link or a [13:27] weight list or a two-actor [13:30] authentication? Like I have all these [13:32] ideas. Um and I couldn't decide until [13:35] the last moment. So I really suggest [13:37] that you decide I think if you create uh [13:41] the login now lovable does that really [13:44] easily for Gmail for Apple just think [13:46] about which emails that you want and you [13:49] know just maybe trying that out before [13:52] the real deal the real implementation [13:56] and then if you want to have a demo you [13:59] can add a demo link that was a really [14:00] cool cool tip from Whitney last time. So [14:04] you can just say here's a demo and then [14:06] you know you can bypass it very easily [14:08] and nowadays level apparently does this [14:11] thing. So whenever uh somebody joins the [14:14] weight list you can um preview the email [14:18] they're going to get. So um once you [14:22] integrate you can also create an email [14:24] within all within lovable and this isn't [14:27] a part of the website apparently. Thanks [14:29] Lovable. And you can even see the [14:34] options with socials, without if a user [14:38] puts in their socials. If they don't put [14:40] in their socials, this is what the email [14:41] is going to look like. Like it's diff [14:43] it's going to be different. Um and then [14:46] you can select it to your own email and [14:50] you will get this this basically this [14:52] previewed email. So it's perfect. And um [14:55] the way to integrate it is like I would [14:58] really suggest using Bravo because Bravo [15:00] is for free and this is what lovable [15:03] suggested me honestly and it's very easy [15:05] to use and just so that you don't have [15:08] to figure all of this out. I'm just [15:10] going to tell it to you there are two [15:11] steps to doing that. Opening an account [15:14] getting an API key from API in the [15:17] settings and then making sure that you [15:20] don't require authorization [15:22] for APIs. That's all because [15:25] automatically it requires some sort of [15:28] um authentication. We want to bypass the [15:30] authentication. You will have that [15:32] option um available to you here in the [15:36] authorized IPs. you want to deactivate [15:38] the API key authorization and you have a [15:42] immediate email um integration with [15:45] Bravo for free for the duration of the [15:49] hackathon or if you want to use it and [15:51] upgrade later probably you can and you [15:54] can test out their emails as such. Um so [15:57] that's pretty much I think all that I [16:00] had planned for you. [16:02] So um I'll open it up for questions. I [16:06] wanted to also spend time if it's [16:08] possible to answer any questions you [16:10] might have. [16:12] >> That was great. Um, thank you so much. [16:15] And you know, you have to share the link [16:17] to women who build. You couldn't just [16:19] tease us with that. We're already I [16:21] know. I was like, wait a minute, what [16:23] what is this? So, make sure [16:26] >> Yeah, it's now offline. It's like uh [16:29] it's a real genuine 8 to 10 people [16:33] dinners right now that we are doing in [16:35] Prague, but if we if there is a wish to [16:38] do it elsewhere, I'm happy to more than [16:40] happy to organize it anywhere else in [16:42] Europe for now. [16:44] >> Well, that's amazing. Um please keep us [16:46] updated on it. Um we have a question [16:50] from Danny. Um the dashboard was [16:53] absolutely great. One question here. [16:55] What uh uh do we need or can or should [16:59] we share a dashboard like that with the [17:01] judges uh like when we submit the [17:03] project or is it more a tool to help us [17:06] visualize the project? [17:08] >> That's a great question. I think um it's [17:12] not necessary uh because I think they're [17:15] looking more at the final product. Um, [17:17] so they're looking at your product and [17:20] understanding how you've implemented [17:21] what you've learned and how much [17:23] research you have done. So I would say [17:26] don't worry about sharing it because if [17:28] you actually do the dashboard, it's [17:30] going to show in your product. [17:34] >> Perfect. And um I can say that she is [17:36] spot on. Um, so we will look at the [17:39] entire project and um, you know, [17:42] anything you have in there is great, but [17:44] we really want to see, you know, what [17:46] you've put together that is moved from [17:48] an idea to a build a an a real product [17:52] and um, anything you've learned along [17:54] the way. So, you put it perfectly. Um, [17:58] would you use your own personal email [18:00] for what you were showing or would that [18:02] be a big no no? [18:05] H well as you've seen I don't have any [18:08] filter in using multiples of demo [18:11] personal slash because I'm a product [18:14] manager I I have I think at this point I [18:17] cannot generate new Gmail accounts [18:19] probably with Google um like I have so [18:21] many test accounts so um I would say [18:25] yeah sure if you have a test account [18:27] already or you want to create it [18:28] beforehand do it beforehand don't lose [18:30] time with it in the hackathon if you [18:32] don't have it just use your own unless [18:34] you have privacy concerns But I don't [18:36] think that um if you put a [18:40] yeah if you put a link to the demo um [18:44] account and fill it automatically so [18:46] people will have access to your Gmail [18:48] address as long as you're okay with that [18:51] um you can place that. [18:53] >> Great. And if you're thinking about any [18:55] of that like what um Deafany showed with [18:57] the Brevo uh you're not able to connect [19:00] to those type of email platforms with [19:02] your personal. It has to be a domain [19:05] that you own. Um, so something to just [19:08] keep in mind there. [19:10] >> Oh, actually, um, with me, I was I was [19:13] able to connect with my own Gmail. So, [19:15] the email itself goes out from my own [19:17] Gmail. [19:18] >> So, there you go. [19:19] >> Um, I I think so, it's like this [19:23] scripted email address, you know, I [19:25] think with my email, but it's [19:28] essentially sending through Bravo, but [19:29] my email's somewhere in there. [19:31] >> Yeah. So, just so you know, [19:33] >> um I'm going to take the qu this next [19:36] question is, "Are there any features [19:37] that may need payments while we build or [19:39] will it be flexible?" So, I'll answer [19:41] that one. Um don't spend money on [19:43] anything during your 48 hours. Save it. [19:47] Um you can set up anything. There's [19:49] usually like a free version out there or [19:52] just set up like a demo. You don't have [19:54] to like please don't spend money. [19:57] Um but here's one right back to Deafany. [20:00] Um, what tools from Anastasia, what [20:02] tools do you suggest for adding [20:04] animations? Um, or does Lovable build [20:07] animations? [20:10] Um, Lovable actually does it for you? I [20:14] think, um, it's a good question to ask [20:18] to the guys from Get to Give because [20:20] they have really cool animations. I [20:22] would I mean, you know, they are [20:24] available um, they are in the alumni [20:27] chat. So feel free to reach out to them [20:29] and ask how did they do these cards [20:30] animations but um to the extent that I [20:34] have explored lovable does add whatever [20:38] animations you want to add. So if you [20:40] say overlap these three cards and then [20:43] make them like flare out, I'm pretty [20:45] sure it's going to be able to do that. [20:47] >> Yeah. [20:49] >> And how easy or difficult was it to [20:51] integrate API keys into the build? [20:54] >> Oh, it's super easy. [20:57] So, um, it even asks you to, um, put in [21:00] the API key. So, if you say to Lovable, [21:03] hey, I want to integrate with Brao, it's [21:06] going to make it as easy for you as [21:08] possible. It even gives you a link to [21:11] Bravo to like sign in and to where to [21:14] get your API key, even like direct deep [21:17] link to wherever you need to go. So once [21:20] you kind of discuss it and plan it with [21:23] lovable, I'm sure that you're just going [21:26] to need to copy the API key and just put [21:28] it in the chat and voila. [21:32] >> Perfect. Um, let's see. Question from [21:37] YouTube. I think I missed it, but how do [21:39] you access the info from the weight [21:40] list? Does it format format it as a CSV [21:43] or a different type of file? [21:47] H that's a good question. Um I think [21:51] it's in your fi in the files of lovable [21:54] so you can see who fills it in. Just ask [21:59] and again like it's something like [22:01] Google it you know nowadays just ask [22:03] lovable it's going to tell you exactly [22:05] where that file is. Um because it might [22:08] be keeping it in different files. I [22:10] think um you know it the program also [22:12] updates really um consistently like [22:16] weekly they are showing you updates. So [22:19] maybe there is um um there are some [22:22] differences between each product or each [22:24] each um each build. So just ask lovable [22:29] I would say. [22:30] >> Great. And uh related to security how do [22:35] you recommend to use lovable to protect [22:37] the project from hackers? [22:41] Yeah. Um, so Lovable, uh, I don't know, [22:44] Marina, if you have the already seen or [22:48] built something with Lovable, but once [22:51] you get into the hackathon, security is [22:53] going to be very much top of mind. Um, [22:56] and Whitney and Elena definitely will [22:59] support you with all the questions in [23:01] terms of the techniques of kind of [23:04] overcoming security issues. And I'm sure [23:06] since our time this has gotten a little [23:09] bit better, but what happens is lovable [23:12] flags each security issue to you [23:14] already. So it detects what might be at [23:17] risk. So a risky situation might be like [23:21] a medium risk where you don't have to [23:23] take even any action. It's just for your [23:25] information. So you will know what are [23:28] the risks associated with the security [23:30] breach or gap that you have in your [23:32] product and what you should do. And you [23:35] can automatically ask Lovable to fix it [23:38] as well, which is just amazing and it's [23:40] just the benefit of it. But definitely [23:43] do I like what what I do is I [23:47] say it in my PRD. I say I want [23:52] um the external [23:54] you know people the visitors who click [23:57] my website to not access any information [24:01] put in by other uh visitors and these [24:05] sort of specific prompts that you will [24:08] put in your PRD will help lovable guide [24:12] around and I think you can also again [24:15] test these out before and do a little [24:19] bit of digging into the channel that we [24:21] are in with participants and alumni [24:24] because this has happened like you know [24:27] time repeats itself this has happened in [24:30] season 2 a lot of us were stuck with [24:32] security issues so I suggest already [24:35] typing in their security checking out [24:37] the errors checking out the how to kind [24:39] of overcome them [24:42] >> absolutely and we do have I think two [24:44] related micro webinars related to [24:46] security this week. One of which is [24:49] actually being hosted by one of the [24:51] winners who had the site that you [24:53] showed. So you everybody can ask about [24:55] those card animations. [24:57] >> Yeah, there you go. They're so helpful. [25:00] Everyone is so helpful. Um ask us [25:02] anything in that chat. We are here for [25:05] you and we're just enjoying the journey [25:07] as as alongside you. [25:10] And uh question from uh Tracy looks like [25:13] if we're setting up something where we [25:15] want to take user information, should we [25:17] expect to bring in a third-party [25:18] integration or can the lovable cloud [25:20] keep that information for you? [25:24] >> Mhm. [25:25] Um I don't know about scaling it. Maybe [25:29] Whitney knows a little bit more than I [25:31] do about lovable cloud. As far as I [25:33] know, yes, Lovable Cloud is pretty great [25:35] at doing just that. But uh Whitney if [25:38] you if there's anything you want to add [25:40] or Xavier go ahead. [25:42] >> Yeah so uh lovable can hold uh levelable [25:44] cloud can uh hold that for you. It is [25:47] secure. It is just like super basease. [25:49] Um and as you grow it can grow alongside [25:52] you. Uh so that is absolutely possible. [25:55] You do have the ability to use superbase [25:58] directly if you want. It's whatever you [26:00] are most comfortable with. [26:03] Um [26:04] >> yeah about the last question I think uh [26:07] I can also answer it with me you can [26:09] too. It should be done in the 48 hours [26:12] uh according to your specific product. [26:14] So once you start building later in the [26:17] game you're going to see the security [26:19] issues show up. [26:21] >> Absolutely. Um so you can start planning [26:25] before Sunday but don't actually start [26:27] building. So you'll want to address the [26:28] security issues as you are building. Um, [26:32] but wow, Deafany, thank you. Um, I [26:35] really love that you touched on so many [26:36] different things because you could tell [26:38] everybody's minds were right there being [26:39] like, "Oh, I want to think about the [26:41] design. I want to think about security." [26:43] So, thank you. And I don't think it was [26:45] a failure. I think that maybe that's [26:48] just me and I don't want anyone to see [26:49] themselves as a failure. [26:51] >> Um, but the fact that you came back and [26:55] are giving to this group is just so [26:58] wonderful. Um, I know that you probably [27:01] looked at season 1 and were like, "Wow, [27:03] how kind of them." I hope you know that [27:05] you have season 3 now looking at you [27:07] that same way. Um, so thank you so much. [27:10] I love that you're in the Discord and [27:13] we're all in this together. So, you're [27:15] amazing. [27:16] >> I am absolutely cheering for everyone. [27:19] I'm sure you guys are going to just kill [27:21] it. It's going to be amazing. and [27:23] whatever you build like you know I think [27:25] we don't give enough uh awards in this [27:29] um hackathon everyone should everyone's [27:32] a winner you know so definitely think [27:35] about it this way I have built as you [27:37] saw like the women who built came out [27:39] after this hackathon and it's just came [27:42] out of my passion of she builds it [27:45] inspired me to do something here locally [27:48] um with my friends with other moms um [27:51] you know how it is you know Last time [27:53] when I was in season 2, I was also in [27:55] the thick of it with my daughter who was [27:58] 2 years old. So, I'm sure there are a [28:00] lot of mothers who are building as they [28:02] take care of their kids. So, definitely [28:05] take a rest when you need, but also it's [28:07] going to be a really fun ride, the 48 [28:10] hours. Um, just give it your all with [28:12] the preparation as well. Believe in [28:14] yourself, trust in yourself that it's [28:17] going to go the way that you want it to [28:20] go. And that's the most important thing. [28:22] and good luck. Enjoy. [28:24] >> Perfectly said. We're going to end with [28:26] that. Thank you so so much and we'll see [28:28] you in the chat.  ## 📺 Video ![](https://www.youtube.com/watch?v=N8W1eI1B_-0) ## Highlights - I was working on the gradient for like literally 20 minutes to get it right. These things could have already been done definitely. - Note: 📚 Pre-sprint brand decisions prevent wasted hackathon hours · - Timestamp ~00:05:30; branding section - Specific, recoverable cost of not pre-committing to brand assets. Applies directly to any time-boxed build context — not just hackathons. **Tags:** #xref #aphorisms **Highlight:** --- · #atom/literature/h1 - My product was for kids — teaching AI to kids — and later on I noticed how hard it is to sell something to kids or even get them to give consent. - Note: 📚 Audience definition before solution prevents unrecoverable product pivots · - Timestamp ~00:09:00; audience research section - Concrete failure case: audience-specific friction (parental consent) only visible after build. The pattern generalizes to any regulated or intermediated user group. **Tags:** #xref #articleseed **Highlight:** --- · #atom/literature/h2 - I didn't spend more than 45 minutes on this and it created me a feature by feature breakdown of what Lovable needs today. - Note: 📚 Lovable can generate a market analysis dashboard from scraped community data in under an hour · - Timestamp ~00:13:30; dashboard section - Establishes a concrete time benchmark and output type (feature breakdown, trust-breaking problems, prioritization view) for no-code PM research workflows. **Tags:** #xref #addendum **Highlight:** --- · #atom/literature/h3 - Encode security intent in the initial prompt, not as a post-build patch - Note: Holds when: the no-code platform (e.g., Lovable) has an active security flagging layer and interprets natural-language security constraints. · ** · #atom/zettel/z1 - Investigate Apify's terms of service for scraping Discord and LinkedIn in commercial/educational contexts — is it within ToS for hackathon audience research? — https://apify.com/terms - Note: ✏️ Fleeting capture #atom/fleeting/f1 - Women Who Build" Prague dinner model — could this be templated as a chapter playbook for other SheBuilds alumni cities? — https://www.youtube.com/watch?v=N8W1eI1B_-0 #fleeting #inbox #storystem - Note: ✏️ Fleeting capture #atom/fleeting/f2 - ** "We as women tend to go easy on ourselves but take what other people are asking for us so seriously." Location: audience research section. This is a broad generalization from personal experience applied to all women; it may alienate participants who do not identify with this pattern. Fix: hedge with "many of us find..." or cite a source. - Note: 🔍 Fallacy: Hasty Generalization #fallacy/hasty-generalization --- # 📝 Full Notes (Deep-Read Analysis) # 🏷️ Prepare for SheBuilds: Before the Hackathon ## 🧭 Table of Contents - [📝 Summary](#-summary) - [📖 Main Content](#-main-content) - [🗒️ Notes](#️-notes) - [📚 References](#-references) - [🗣️ Paraphrase](#️-paraphrase) - [📰 Newsletter](#-newsletter) - [✅ Action Items](#-action-items) - [🧪 Hypotheses](#-hypotheses) - [🎯 Conclusion](#-conclusion) - [🔄 Refactor Appendix](#-refactor-appendix) --- ## 📝 Summary - **What happened:** Deafany (Stephanie), a Season 2 SheBuilds alumna and product manager, delivered a pre-hackathon coaching session to Season 3 participants covering branding, audience research, dashboard building, animations, and API integrations using the Lovable no-code platform. - **Why it matters:** Participants have 48 hours to build a real product. Preparation done *before* the clock starts can be the difference between shipping a polished product and getting stuck on logos and login flows during the sprint. - **Key stakeholders:** SheBuilds Season 3 participants, program leads (Whitney, Elena, Xavier), Lovable platform team, alumni mentors. - **Risks and tradeoffs:** Spending precious hackathon hours on branding, API setup, or security issues that could have been resolved in advance; the temptation to over-engineer (gradient logos, undecided auth flows) instead of focusing on validated audience problems. - **Bottom line:** Do your brand, audience research, integration testing, and animation planning *before* the hackathon. Treat it like a real work project. Failure is a data point, not an identity. --- ## 📖 Main Content ### 🎯 Core Argument Deafany's central thesis is that hackathon success is front-loaded: the preparation window before the 48-hour sprint determines product quality more than raw coding hours. She frames this through her own Season 2 experience — she did not win, but she extracted enough lessons to advise the winners, grow as a PM, and launch a women's community in Prague. ### 🧠 Think From Your Future Winning Self She opens with a mental model: imagine you have already won. What would that future self tell you to do *today*? This reframes mundane prep tasks (create social links, set up a LinkedIn, generate brand assets) as strategic, not administrative. ### 🎨 Step 1 — Branding Before the Sprint - Choose three main brand colors and a font *now*, before the hackathon starts. - Load those exact values into Lovable via explicit prompts; the platform can copy colors/fonts inconsistently if not specified. - Use Canva to create a brand kit; do not improvise gradients live during the sprint (she lost 20 minutes to this). ### 👥 Step 2 — Audience Before Solution - Resist the impulse to start with a solution. Pick a community you are already passionate about. - Run even a minimal survey — three conversations can surface patterns. - Scrape competitor Discord channels, Reddit threads, and LinkedIn using tools like **Apify** (free, community-built scrapers shared on Reddit) to understand what audiences are already saying. - Her own product (teaching AI to kids) hit a wall because she underestimated the friction of parental consent and selling to minors — an audience insight she could have caught earlier. ### 📊 Step 3 — Build a Market Analysis Dashboard in Lovable - Before writing a PRD, build a standalone research dashboard in Lovable (she built hers in under 45 minutes). - Her example: she scraped Lovable's own Discord feedback channel and produced a feature-by-feature breakdown, a trust-breaking problems analysis, and a prioritization view — all inside Lovable. - This replaces the instinct to jump straight to PRD writing when you do not yet have deep market context. ### ✨ Step 4 — Plan Animations Early - Animations are underrated in hackathon judging; judges visit your website and visual delight matters. - Lovable does not add animations by default — you must prompt for them explicitly (hover effects, magnetic scroll, 3D card fly-outs). - Reference: the Season 2 winner's site "Get to Give" uses 3D card animations that feel subtle but are highly polished. ### 🔌 Step 5 — API and Integration Decisions Before Day One - Decide your authentication flow (Gmail login, magic link, waitlist, demo bypass) *before* the sprint starts. - **Brevo** (formerly Sendinblue) is Deafany's recommended free email integration; Lovable natively suggests it and walks you through obtaining the API key. - Two-step Brevo setup: (1) get API key from settings, (2) deactivate API key authorization in Authorized IPs so Lovable can connect without extra authentication overhead. - Lovable can generate and preview onboarding emails — including conditional variants based on whether the user provided social links. - Waitlist data is stored in Lovable's files; ask Lovable directly where to find it or how to export it. ### 🛡️ Security - Lovable flags security issues automatically with severity levels (medium, high, etc.) and offers to fix them. - Best practice: write security intent directly into your PRD prompt, e.g., "visitors should not be able to access data entered by other users." - Security issues tend to appear as you build — they cannot be fully pre-solved, but anticipating them reduces panic during the sprint. - SheBuilds is hosting dedicated security micro-webinars during the week. ### 💬 Q&A Highlights - **Should you share the research dashboard with judges?** No — the product itself should reflect your research. The dashboard is a tool, not a deliverable. - **Personal email for testing?** Fine if you have no privacy concerns; create a test account beforehand to avoid losing sprint time. - **Brevo + personal email:** Brevo can send from your personal Gmail address (email is routed through Brevo but the sender address is yours). - **Spend money during the hackathon?** No — use free tiers and demo modes throughout. - **Lovable Cloud for user data:** Lovable Cloud (backed by Supabase) can securely store user data and scale; you can also connect directly to Supabase if preferred. --- ## 🗒️ Notes #### ⏱️ [00:00:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=0) Whitney introduces Deafany - Frames the session as alumni giving forward. - Sets tone: failure is acceptable, learning is the point. #### ⏱️ [00:01:30](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=90) Deafany opens — Season 2 failure framing - Did not win Season 2; reached out to winners post-hackathon. - Framing: "this will not be your last product" — reusable career perspective. - Mentions building Women Who Build (Prague dinner community) as a direct outcome of SheBuilds inspiration. #### ⏱️ [00:03:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=180) Future-self mental model - "What would your future winning self tell you to do today?" - Specific examples: social links, LinkedIn profile, brand assets, AI-generated video content (Grok mentioned for video). #### ⏱️ [00:05:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=300) Step 1 — Branding - Three colors + one font in Canva before the sprint. - Explicit Lovable prompts for exact font/color reproduction. - Lost 20 minutes to gradient logo during Season 2 — cautionary tale. #### ⏱️ [00:08:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=480) Step 2 — Audience research - Pick community first, solution second. - Kids product failure: parental consent friction not anticipated. - Tools mentioned: Apify, Reddit scrapers, Discord scraping. - Coach insight ($1,000/month): survey people about their *day*, not your solution. - Gender observation: women tend to take others' requests seriously but go easy on themselves — apply work-project rigor to personal projects. #### ⏱️ [00:13:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=780) Step 3 — Market analysis dashboard in Lovable - Built in under 45 minutes using scraped Discord data. - Outputs: feature breakdown, trust-breaking problems, prioritization view. - Replaces premature PRD writing for participants who lack deep market context. #### ⏱️ [00:17:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=1020) Step 4 — Animations - Hover effects, magnetic scroll, 3D card animations. - Lovable can execute these — but only with explicit prompts. - Get to Give (Season 2 winner) cited as animation reference; team is in alumni Discord. #### ⏱️ [00:20:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=1200) Step 5 — API integrations and auth - Decide login method before the sprint: Gmail OAuth, magic link, waitlist, demo bypass. - Brevo: free, Lovable-recommended, sends from personal Gmail. - Two setup steps: API key + deactivate authorization requirement. - Lovable generates and previews conditional onboarding emails natively. #### ⏱️ [00:26:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=1560) Q&A — dashboard, emails, security, data storage - Research dashboard: tool for builder, not a judge submission. - Security: Lovable auto-flags issues; encode intent in PRD prompts. - Lovable Cloud / Supabase: viable for user data storage, scalable. - Waitlist data: stored in Lovable files; ask Lovable for location. #### ⏱️ [00:35:00](https://www.youtube.com/watch?v=N8W1eI1B_-0?t=2100) Closing — mindset - "Everyone's a winner" — reframes the 48 hours as a learning artifact regardless of placement. - Rest when needed; give full effort in preparation. - Women Who Build community in Prague mentioned as post-hackathon outcome. --- ## 📚 References - [Lovable](https://lovable.dev) — no-code AI app builder used throughout the hackathon - [Brevo (formerly Sendinblue)](https://brevo.com) — free email integration recommended for waitlist/onboarding emails - [Apify](https://apify.com) — web scraping platform for Discord, Reddit, LinkedIn audience research - [Canva](https://canva.com) — brand kit creation (colors, fonts, assets) - [Supabase](https://supabase.com) — backend database underlying Lovable Cloud - [Get to Give](https://gettogive.com) — Season 2 SheBuilds winner, cited for animation quality - [Grok](https://grok.x.ai) — mentioned for AI video/content generation --- ## 🗣️ Paraphrase - Deafany's core message is that 48-hour hackathons are won or lost in the days before they start. Participants who arrive with a defined brand, a validated audience, a working API integration, and a clear animation plan are able to spend their sprint hours actually building instead of making foundational decisions under time pressure. - The secondary message is a reframe of failure. Not winning Season 2 was not the end of Deafany's product journey — it was the beginning of her growth as a PM, an independent community builder, and a mentor. The hackathon's real product is the participant, not the submission. --- ## 📰 Newsletter **🛠️ Build Smarter, Not Faster: Pre-Hackathon Prep Lessons from SheBuilds** If you've ever sat down to build something and lost an hour to picking the right shade of blue — this one's for you. Deafany, a Season 2 SheBuilds alumna and product manager, shared a refreshingly practical session with Season 3 participants: the best 48-hour hackathon starts days before the timer begins. Her checklist? Lock in your brand kit (three colors, one font — in Canva, before day one). Choose your *audience* before your solution. Build a market research dashboard in Lovable using scraped Discord or Reddit data from competitor communities. Decide your auth flow in advance. And don't forget animations — judges visit your website, and Lovable won't add hover effects unless you ask. Her tool stack for the prep phase: **Canva** for brand assets, **Apify** for audience scraping, **Brevo** for free email integrations, and **Lovable** for everything else (including a 45-minute market analysis dashboard she built live). The deeper lesson: treat your side project with the same rigor you'd apply to a work deliverable. And if you don't win? You've still shipped, learned, and — in Deafany's case — launched a women's dinner community in Prague. Not bad for a "failure." 🔗 [Watch the full session](https://www.youtube.com/watch?v=N8W1eI1B_-0) --- ## ✅ Action Items #### DevRel - Create a pre-hackathon prep checklist (brand, audience, integrations, animations) as a reusable resource for future SheBuilds cohorts. - Develop a short-form "market dashboard in Lovable" tutorial based on Deafany's 45-minute workflow. - Publish a Brevo + Lovable integration quickstart guide with the two-step API key setup documented. #### Community - **Contributors:** Pin Deafany's session recap and key tool links (Apify, Brevo, Canva, Get to Give) in the Season 3 Discord. - **Maintainers:** Invite Get to Give winners to host an animation-focused micro-session; they are already flagged as alumni in Discord. - **Sponsors:** Document the "Women Who Build" offline community model (Prague dinners) as a potential replicable chapter format for SheBuilds expansion. #### WordPress - **Training:** The audience-research-before-solution framework (survey → scrape → pattern → PRD) is transferable to WordPress plugin development and theme design; consider a lesson plan adaptation. - **Community:** The women-in-tech focus and "give forward" alumni model maps well to WordCamp mentorship tracks. #### The WPCC - **Programs:** Deafany's session demonstrates high value from structured alumni-to-cohort mentorship. Consider a formalized "alumni mentor" track with a defined session format and follow-up loop. --- ## 🧪 Hypotheses 1. **Preparation quality predicts sprint output quality:** Participants who complete brand, audience research, and integration setup before the hackathon will produce higher-rated products than those who begin cold on day one. 2. **Audience-first framing reduces pivot friction:** Teams that define their target community before their solution will require fewer mid-sprint pivots than solution-first teams. 3. **Explicit animation prompting improves judge scores:** Products that include intentional hover/scroll animations (explicitly prompted into Lovable) will receive higher design scores from judges than default Lovable outputs. 4. **Research dashboards accelerate PRD writing:** Participants who build a Lovable market analysis dashboard pre-sprint will write higher-quality PRDs in less time during the sprint. 5. **Auth flow pre-commitment reduces lost hours:** Participants who decide on their authentication approach (Gmail OAuth vs. magic link vs. waitlist) before day one will lose fewer sprint hours to login implementation indecision. 6. **Alumni failure narratives increase participant resilience:** Cohorts that hear structured failure-and-learning narratives from alumni will report higher psychological safety and willingness to pivot during the sprint. ### 📊 Methodology - **H1:** Retrospective survey of Season 3 participants comparing prep hours logged against judge scores and self-reported product completeness. - **H2:** Coded interview analysis of teams that pivoted mid-sprint vs. those that did not, correlating with pre-sprint audience definition documentation. - **H3:** Blind judging rubric analysis scoring animation presence against overall design score across all submitted products. - **H4:** Compare PRD word count, specificity, and time-to-first-build between dashboard-builders and non-dashboard-builders. - **H5:** Sprint time log analysis: minutes spent on auth implementation, correlated with pre-sprint auth decision documentation. - **H6:** Pre/post psychological safety survey (adapted Edmondson scale) across cohorts that received vs. did not receive alumni failure narratives. ### ❓ Questions - Does Lovable's security flagging system cover OWASP Top 10 categories, or only platform-specific vulnerabilities? - At what audience size does Lovable Cloud / Supabase require a paid tier, and what is the ceiling for free hackathon use? - Are Apify's community-built scrapers for Discord and LinkedIn within those platforms' terms of service for hackathon/research use? - What specific judging rubric does SheBuilds use, and how much weight is given to design/animation vs. problem-solution fit? - How does "Women Who Build" (Prague) plan to scale — and is there an open-source playbook for starting a chapter? ### 🔮 Future Research - Longitudinal study: do SheBuilds participants who do not place still go on to ship follow-on products at higher rates than comparable no-code learners? - Comparative analysis of winning SheBuilds submissions: what structural features (auth, animations, data layer, audience specificity) correlate with placement? - Survey of PM practitioners: how widely is the "build a research dashboard before the PRD" practice used outside hackathon contexts? - Effectiveness of scraping-based audience research (Apify/Discord) vs. direct user interviews for early-stage product discovery. ### 💡 Implications - **Practice:** The pre-hackathon checklist model could be standardized across accelerator programs and no-code bootcamps as a replicable onboarding artifact. - **Education:** The "future self" mental model and the "work-project rigor" reframe are teachable heuristics for reducing imposter syndrome in non-traditional tech entrants. - **Security:** Encoding security intent into PRD prompts (rather than patching post-build) suggests a prompt engineering best practice for no-code security hygiene. - **Funding models:** If alumni-led mentorship is this high quality, it represents a zero-cost program asset worth structuring formally (mentorship stipends, session recordings, reusable curricula). --- ### 🧠 Learning Styles Assessment - **Reading level estimate:** Flesch-Kincaid Grade Level ≈ 8–9. Conversational, accessible, light on jargon. Appropriate for a broad adult audience with no PM or technical background required. #### 👀 Visual learners - **Present:** Deafany references her Lovable dashboard and animation examples verbally; a website (Get to Give) is cited. - **Improve:** Screenshots of the market analysis dashboard, Brevo API settings screen, Canva brand kit template, and Get to Give animation examples would significantly increase comprehension. #### 🔊 Aural learners - **Present:** Full video with spoken delivery; conversational tone aids listening. - **Improve:** A 3–5 minute audio summary of the five preparation steps would serve podcast-style learners who want a refresher without re-watching. #### 🗣️ Verbal learners - **Present:** Rich in analogies (iceberg, future self, work-project framing), storytelling, and concrete examples. - **Improve:** A glossary of terms (Brevo, Apify, Supabase, magic link, PRD) for participants new to the stack would reduce friction. #### 🧩 Kinesthetic learners - **Present:** Step-by-step instructions for Brevo setup; live demo references. - **Improve:** A downloadable pre-hackathon checklist with checkbox tasks (create Canva brand kit, get Brevo API key, run Apify scrape) would be immediately actionable. #### 🧠 Logical learners - **Present:** Clear five-step structure with causal reasoning (if you do X before the sprint, Y will not cost you time during it). - **Improve:** A decision tree for auth flow selection (Gmail OAuth vs. magic link vs. waitlist vs. demo) and a risk/effort matrix for the five prep steps would sharpen the logical scaffolding. #### 🧑‍🤝‍🧑 Social learners - **Present:** Live Q&A, Discord community references, alumni chat callout. - **Improve:** Structured peer pairing (e.g., "find an accountability partner and share your brand kit by Friday") would create a social accountability layer. #### 🧘‍♀️ Solitary learners - **Present:** The "future self" exercise is inherently self-reflective. - **Improve:** A set of journaling prompts (Who is my audience? What would my winning self have done differently? What is my auth strategy?) would scaffold individual reflection. #### 🌿 Naturalistic learners - **Present:** Not explicitly addressed. - **Improve:** Framing the product lifecycle (idea → build → pivot → product 2) as an ecosystem / growth cycle could resonate with naturalistic thinkers. --- ### 📊 Accessibility Assessment #### Screen reader semantics - **Present:** Video format; YouTube auto-captions available. - **Improve:** Captions should be human-reviewed (auto-captions misrender "Brevo" and "Apify"); descriptive alt text for any slides or screenshots shared. #### Keyboard navigation - **Present:** YouTube player is keyboard navigable. - **Improve:** Any linked companion resources (checklists, dashboards) should be tested for keyboard reachability. #### Contrast and readability - **Present:** Not assessable from transcript alone; slides were referenced but not visible in transcript. - **Improve:** Ensure any published slide deck or resource page meets 4.5:1 contrast minimum. #### Text resizing and reflow - **Present:** N/A for video. - **Improve:** Any companion web resource should be tested at 200% zoom. #### Media and captions - **Present:** YouTube auto-captions exist. - **Improve:** A downloadable transcript (cleaned) and audio description of any screen-shared dashboards or animations would serve deaf and low-vision participants. #### Forms and interactions - **Present:** N/A for the video itself. - **Improve:** Any Lovable-built products participants create should follow the accessible forms guidance Deafany alludes to (error messages, label clarity). #### Standards reference - Gaps map to WCAG 2.2 AA: 1.2.2 (captions), 1.2.5 (audio description), 1.4.3 (contrast). --- ### 🌍 DEIB Assessment #### Cultural breadth - **Present:** Prague-based offline community mentioned; implicit global reach via Discord. - **Improve:** Explicit acknowledgment that hackathon participation spans time zones and that prep steps should be adaptable to participants with limited internet access (e.g., lighter-weight alternatives to Apify scraping). #### Inclusive language - **Present:** Gender-neutral for most content; "women" used deliberately in the program context, which is appropriate and intentional. - **Improve:** The observation that "women tend to go easy on themselves" is a generalization that may not resonate universally; framing as "many of us" or citing the observation as a pattern rather than a rule would be more inclusive. #### Neurodiversity support - **Present:** Clear five-step structure aids sequential processing; numbered steps reduce cognitive load. - **Improve:** Explicit "TL;DR" callouts per step and consistent heading use in any written companion would support ADHD and dyslexic participants. #### Representation - **Present:** Female PM and community builder in a technical role is meaningful representation. - **Improve:** The session could explicitly acknowledge participants from non-PM, non-tech backgrounds (e.g., teachers, caregivers, artists) who are equally valid builders. #### Psychological safety - **Present:** Strong — the failure framing, "everyone's a winner" close, and explicit acknowledgment of mothers building while caregiving create a psychologically safe container. - **Improve:** A clear code of conduct reference for the Discord and a moderated Q&A structure would formalize safety beyond tone. --- ### ✅ Logical Validity #### Strong reasoning - **Future-self heuristic (throughout):** A well-established implementation intention strategy in behavioral psychology. Framing prep tasks as actions your successful future self would have taken is a sound motivational device. Strong because it is directional and specific. - **Audience-before-solution argument (Step 2):** Deductive and evidence-backed by her own product failure (kids/parental consent). The causal chain (undefined audience → unexpected friction → failed product) is coherent and falsifiable. - **Pre-sprint integration decisions (Step 5):** Inductive reasoning from observed pattern (auth indecision costs sprint time) generalized as a rule. Sound, with clear actionable output. #### Fallacies - **Hasty generalization — gender observation:** "We as women tend to go easy on ourselves but take what other people are asking for us so seriously." Location: audience research section. This is a broad generalization from personal experience applied to all women; it may alienate participants who do not identify with this pattern. Fix: hedge with "many of us find..." or cite a source. - **Anecdotal evidence — future-product claim:** "This will not be the last product you build." Deafany presents her own trajectory as the expected norm. Location: opening remarks. Fix: frame as a possibility ("for many of you, this will be the first of many products") rather than a certainty. #### Tone of voice - **Warm, encouraging, peer-to-peer.** Examples: "I'm absolutely cheering for everyone," "whatever you build — you know — I think we don't give enough awards." The tone reduces hierarchy between mentor and participant, which is appropriate for the psychological safety goal of the session. --- ### 🛡️ DARVO Analysis - **Deny:** Not applicable. This session contains no defensive behavior; Deafany proactively acknowledges her own failure. - **Attack:** Not applicable. No adversarial dynamics present. - **Reverse victim/offender:** Not applicable. - **Note:** DARVO is not a relevant frame for this content. The session is a collaborative, alumni-led coaching context with high psychological safety and no contested claims. The failure narrative is voluntarily disclosed and constructively framed. --- ## 🎯 Conclusion Deafany's pre-hackathon session delivers a compact, high-signal preparation framework: brand before day one, audience before solution, research dashboard before PRD, explicit animation prompts, and auth decisions locked in advance. The meta-lesson is methodological rigor applied to personal projects — the same standards used in professional PM work. **Practical next steps for participants:** 1. Open Canva today; define three colors and one font. 2. Identify three people in your target community and send a survey this week. 3. Scrape a competitor Discord or Reddit thread with Apify; paste results into Lovable and generate a market analysis dashboard. 4. Open a Brevo account, get your API key, deactivate authorization, and test the integration before Sunday. 5. Prompt Lovable for at least one hover animation and one scroll effect before the sprint ends. 6. Write your security intent into your first PRD prompt ("visitors should not see other visitors' data"). --- ## 🔄 Refactor Appendix — Prepare for SheBuilds: Before the Hackathon ### 📚 Literature Atoms #### [H1] Pre-sprint brand decisions prevent wasted hackathon hours > [!quote] ID:rw_shebuilds_001 > "I was working on the gradient for like literally 20 minutes to get it right. These things could have already been done definitely." **Context & use:** - Timestamp ~00:05:30; branding section - Specific, recoverable cost of not pre-committing to brand assets. Applies directly to any time-boxed build context — not just hackathons. **Tags:** #xref #aphorisms **Highlight:** [view highlight](readwise://shebuilds_001)  --- #### [H2] Audience definition before solution prevents unrecoverable product pivots > [!quote] ID:rw_shebuilds_002 > "My product was for kids — teaching AI to kids — and later on I noticed how hard it is to sell something to kids or even get them to give consent." **Context & use:** - Timestamp ~00:09:00; audience research section - Concrete failure case: audience-specific friction (parental consent) only visible after build. The pattern generalizes to any regulated or intermediated user group. **Tags:** #xref #articleseed **Highlight:** [view highlight](readwise://shebuilds_002)  --- #### [H3] Lovable can generate a market analysis dashboard from scraped community data in under an hour > [!quote] ID:rw_shebuilds_003 > "I didn't spend more than 45 minutes on this and it created me a feature by feature breakdown of what Lovable needs today." **Context & use:** - Timestamp ~00:13:30; dashboard section - Establishes a concrete time benchmark and output type (feature breakdown, trust-breaking problems, prioritization view) for no-code PM research workflows. **Tags:** #xref #addendum **Highlight:** [view highlight](readwise://shebuilds_003)  --- ### 🧠 Permanent Candidates #### [P1] Encode security intent in the initial prompt, not as a post-build patch **Claim** (one sentence, in my voice): Security requirements specified in a no-code PRD prompt produce fewer vulnerabilities than security treated as a cleanup step after the build is complete. **Evidence pointer:** [[#H1 Pre-sprint brand decisions prevent wasted hackathon hours]] — security intent follows the same logic as brand decisions: pre-commitment reduces remediation cost. **Scope & conditions:** - Holds when: the no-code platform (e.g., Lovable) has an active security flagging layer and interprets natural-language security constraints. - Breaks when: the platform does not parse intent-level security prompts, or when the threat model requires formal security review beyond prompt-level controls. **Links:** - [[No-Code Security Patterns]] — because extends - [[PRD Writing Best Practices]] — because applies-in product management contexts - [[Lovable Platform Capabilities]] — because evidence-for  --- ### ✏️ Fleeting Queue - [F1] Investigate Apify's terms of service for scraping Discord and LinkedIn in commercial/educational contexts — is it within ToS for hackathon audience research? — https://apify.com/terms #fleeting #inbox - [F2] "Women Who Build" Prague dinner model — could this be templated as a chapter playbook for other SheBuilds alumni cities? — https://www.youtube.com/watch?v=N8W1eI1B_-0 #fleeting #inbox #storystem ### → Things 3 (max 2 actions) - [ ] Create a Canva brand kit template pre-populated with three-color + one-font slots for hackathon onboarding — share in SheBuilds Discord #things - [ ] Draft a Brevo + Lovable integration quickstart (two-step API key setup) as a standalone resource for Season 3 participants #things --- ``` Processed: Prepare for SheBuilds — Before the Hackathon - Literature atoms: 3 / 5 cap - Permanent candidates: 1 / 3 cap - Fleeting items: 2 - Things 3 actions: 2 / 2 cap - Konik tags used: #xref, #aphorisms, #articleseed, #addendum - Triage callouts populated: Cross Reference, Article Ideas, Vocabulary (empty) - Quality gates: ✓ verbatim quotes present / ✓ atomicity markers present / ✓ link annotations with relationship types / ✓ no [MISSING] strings / ✓ tagged highlights routed to callouts / ✓ view highlight links present - Override conditions triggered: none ``` --- # 📄 Full Transcript *Auto-generated from YouTube captions · timestamps roughly every 30 seconds · [watch on YouTube](https://www.youtube.com/watch?v=N8W1eI1B_-0)* **[00:00]** far. Um, and it's only the beginning. It's uh it's like a an iceberg, right? We're right at the tip and we're going to keep on going. And it's all thanks to our amazing alumni. Uh, one of whom is here, of course, Stephanie from uh season 2. And so, I'm going to shut up and let her do her thing because I know she's going to bestow upon you all so much more wisdom. Um, and so please, you know, join me in thanking you, uh, **[00:30]** Deafany for taking the time, wanting to share this, to give forward to this next season. Um, we really, really appreciate your time. >> Yeah, sounds great. Thank you, Whitney. I'm super excited to be here with all of you. So, hi everyone. Um, it was such a thrill to join and be a part of season 2. I remember getting accepted and it was so exciting to meet everyone to share all the ideas. So, I'm sure you're having a thrill and you will learn so **[01:03]** much in this journey. So, I'm sure that Elena and Whitney and Xavier have already said this to you multiple times. But definitely look at this as a opt opportunity to learn and whatever happens even if you fail that's the part of the journey without failure there won't be product number two product number three and I can reassure you that for sure this will not be the last product that you build it will not be your last pivot you will pivot multiple **[01:34]** times after this and build multiples of products with lovable uh at least that's what happened to And I've saw that some of your messages, some of your building communities, etc. These are great way to start understanding your audience and then turn it into a product. So I think you all are ready for it and I'm so excited to see what comes out of it. And today I'm going to share with you what I have done as a failure story. So I didn't win season two, but I learned so much from **[02:06]** it. I in the end reach out to the winners as well to ask them what did they do differently. So I'm going to share a little bit of those as well. I'm going to share a little bit of what I think I do differently as a product manager now today. You know looking at back at my work. I have grown so much since then in my in my job and also externally. um you know doing some workshops, speaking etc and engaging with the community of women has been absolutely my passion since then. So **[02:38]** let's get started and I'll share um sort of a few steps that you can take before the hackathon actually starts and really practical steps. Um some of them are going to be technical so maybe we can even look at that how to integrate things as well. So um first part is again as I said you should be thinking about your self as if you have won already and what would this **[03:09]** self do best and what would they come back from the end of season 3 let's say you won and you delivered this amazing project amazing product what would yourself future self tell you today to hey get this done today. Think about those. Is it um creating social media links already? Is it creating your LinkedIn account so that you can post later on? That also creates a little bit of validity. You can then put those links into your product as well. Maybe **[03:41]** it's creating the whole brand assets, brand line. Maybe generating some videos using AI like Grock is really great at videos and video content. Um like one of my personal learnings has been always looking back and saying like these little details, these things I could have done earlier and my screen share has stopped but we are still here. No worries. And so basically **[04:13]** um I'm going to give you these little technical tips about um what I do differently and I think you should take a look at. So number one is branding. So we're going to talk a little bit about branding. Um how to think about your audience, how you can use lovable to create a dashboard about your whole market analysis, and then looking a little bit more into the details. So I think one of the things I really lived through was, oh, okay, I don't **[04:45]** have my logo exactly. Let me put a gradient into the logo because I think that's going to look cooler. And then working on the gradient for like literally 20 minutes to get it right. These things could have already been done definitely. Um so think about the colors. Um I was not really thinking in detail. I think I could do much better. Um I was just thinking like oh this will go well you know and oh I like this other website. I think this will this **[05:16]** lets me like just copy paste that, screenshot that because that's what I knew. But I think going into Canva and creating a nice little model that really is aesthetic for you, having three main colors, this is going to really help you uh grow the brand. Have a font already and uh ask Lovable to exactly put that font in because it can sometimes copy differently. So make sure like it is the exact font. ask it to double check the color um etc. So have that brand exist **[05:47]** before you know the week already starts. Um step number two I think picking the uh the audience before you uh go to the solution is a really important one because we often try to start with the solution. And for me what was really key was going back later after the hackathon and being like um cuz my product was for kids. It was for um teaching AI to kids and later on I noticed that how hard it **[06:20]** is to sell something to kids and or even get them to you know um give consent about a lot of things. You need to involve parents. So really picking your community that you are passionate about already today. I think you're allowed to pivot slightly. So you can maybe already think about that if you can um or if this is the thing you want to go for and making sure you listen to the idea and listen to the audience and where they **[06:52]** are and finding out what they are thinking. You know what are their problems? what do they need exactly? And that's going to help you refine your even if you have the idea and the solution, that's going to really be the key. Um, so for example, I work with a coach for $1,000 a month for building a business. And what she told me essentially was and where I got blocked before was really making a survey asking **[07:24]** people what they want, what are their problems, what how can they resolve them or how do they imagine resolving them or even not even going to the solution. How do they go about their day about you know your product about the the industry that you are thinking about? And this is going to give you so many ideas and so much patterns that going to come out if you talk to just three people even. And um when you once you find where they hang out, you can also use some third **[07:55]** party tools like Appify or there are some third party tools built by you know engineers all around the world shared across Reddit for free to use to scrape um social media, scrape LinkedIn, scrape uh Discord channels of products of competitors to understand what the audiences are talking about. So, I would go ahead and do that already today. Really see what people are talking about. Do it as if you're doing it for your work because I think often times we **[08:28]** also as women, we tend to go easy on ourselves, but we take what other people are asking for us so seriously because we have this compassion to serve and you know this intrinsic empathy to help others. So, it can be really hard to help ourselves. So, I really ask you to take this project as seriously as you would a work project and really hone down on how would I go about this at my work at my job. **[09:00]** And that brings me to my next point. I think um what I have done for a lot of other projects for work for interviews in product management in tech u build a standalone dashboard. This is going to be your um let's say resource and you can look at that uh and I will show you one that I have done and just as I said I have scraped uh lovable and lovable uh feedback channel actually for the **[09:31]** lovable product and um lovable helped me create um I didn't spend more than 45 minutes on this and uh it created me a feature by feature breakdown of what lovable needs today. Um there's a deep analysis into trustbreaking problems of the audience that are writing actively onto discord channel about what's missing about lovable and it also gave me a prioritization dashboard beautiful like what's missing what do the **[10:01]** community say and this is so useful so I think um if you you know if I were a product manager this would be absolutely amazing to decide what feature I'm building next and I I think also as product managers we tend to just directly go to PRD writing or we hear or if we're in tech as well we just imagine okay I'm just going to write the PRD there is a lot of strategy that goes into the work that happens before writing a PRD because if you're a **[10:32]** feature PM often you're aware of the market you're aware of the customer issues you're in it you know you're in the thick of it so you know most of the competitors what they're doing like yesterday what they've done. So you can write the PRD tomorrow but in this case you are just starting this out probably and you need all the information all the details you can have and lovable is actually offering this to you already. So definitely pull this **[11:04]** out, create something like this and then later on I even recommend that you create um one example of a prototype potentially and see how you know lovable works. See how you can um change different things, how you can add integrations. So try things out that you plan to use later on. And that brings me to my next um next point. **[11:34]** Last but not least, think about the animations and the cool tricks that you're going to have that are going to be attention grabbing for well, not only the people and your audience, but this is a hackathon. The jury and the judges are going to come into your website. So, how can you make it really cool? And I think one of the very um underrated things is animations. You can animate at a hover. You know, it can pop up. Like **[12:05]** my slide doesn't do this. But you know, elsewhere, for example, this little website that I've built for my offline community here in Prague, we do some dinner and drinks with women. And you see, I did some little um animations at Hover. And these Lovable doesn't do by default. So, I'm just kind of putting this out there that you might be like, "Oh, I already know this." Um, just think about what kind of animations are going to be useful to you. It might be a magnetic scroll that once you scroll, it locks in. It might be, and **[12:37]** here's a really great example from the winner of last year, last season. Maybe you're aware of this website. It's called Get to Give and it's so beautiful. And they have really cool animations. like you don't even notice that when it happens, but these cards come out and everything's kind of 3D moving up and down. So, you can definitely ask Lovable to do these things, but it's just not going to give it to you for free. So, just a little reminder, think about these things. And **[13:07]** then my last point is going to be let's take a look at how we integrate with API keys and APIs and lovable because this is also something that I played a lot like back and forth in my head. Oh, am I going to have a Gmail login? Am I going to have a like a a magic link or a weight list or a two-actor authentication? Like I have all these ideas. Um and I couldn't decide until the last moment. So I really suggest **[13:37]** that you decide I think if you create uh the login now lovable does that really easily for Gmail for Apple just think about which emails that you want and you know just maybe trying that out before the real deal the real implementation and then if you want to have a demo you can add a demo link that was a really cool cool tip from Whitney last time. So you can just say here's a demo and then you know you can bypass it very easily **[14:08]** and nowadays level apparently does this thing. So whenever uh somebody joins the weight list you can um preview the email they're going to get. So um once you integrate you can also create an email within all within lovable and this isn't a part of the website apparently. Thanks Lovable. And you can even see the options with socials, without if a user puts in their socials. If they don't put **[14:40]** in their socials, this is what the email is going to look like. Like it's diff it's going to be different. Um and then you can select it to your own email and you will get this this basically this previewed email. So it's perfect. And um the way to integrate it is like I would really suggest using Bravo because Bravo is for free and this is what lovable suggested me honestly and it's very easy to use and just so that you don't have to figure all of this out. I'm just **[15:10]** going to tell it to you there are two steps to doing that. Opening an account getting an API key from API in the settings and then making sure that you don't require authorization for APIs. That's all because automatically it requires some sort of um authentication. We want to bypass the authentication. You will have that option um available to you here in the authorized IPs. you want to deactivate the API key authorization and you have a **[15:42]** immediate email um integration with Bravo for free for the duration of the hackathon or if you want to use it and upgrade later probably you can and you can test out their emails as such. Um so that's pretty much I think all that I had planned for you. So um I'll open it up for questions. I wanted to also spend time if it's possible to answer any questions you might have. **[16:12]** >> That was great. Um, thank you so much. And you know, you have to share the link to women who build. You couldn't just tease us with that. We're already I know. I was like, wait a minute, what what is this? So, [laughter] make sure >> Yeah, it's now offline. It's like uh it's a real genuine 8 to 10 people dinners right now that we are doing in Prague, but if we if there is a wish to do it elsewhere, I'm happy to more than happy to organize it anywhere else in **[16:42]** Europe for now. >> Well, that's amazing. Um please keep us updated on it. Um we have a question from Danny. Um the dashboard was absolutely great. One question here. What uh uh do we need or can or should we share a dashboard like that with the judges uh like when we submit the project or is it more a tool to help us visualize the project? >> That's a great question. I think um it's not necessary uh because I think they're **[17:15]** looking more at the final product. Um, so they're looking at your product and understanding how you've implemented what you've learned and how much research you have done. So I would say don't worry about sharing it because if you actually do the dashboard, it's going to show in your product. >> Perfect. And um I can say that she is spot on. Um, so we will look at the entire project and um, you know, anything you have in there is great, but we really want to see, you know, what **[17:46]** you've put together that is moved from an idea to a build a an a real product and um, anything you've learned along the way. So, you put it perfectly. Um, would you use your own personal email for what you were showing or would that be a big no no? H well as you've seen I don't have any filter in using multiples of demo personal slash because I'm a product manager I I have I think at this point I **[18:17]** cannot generate new Gmail accounts probably with Google um like I have so many test accounts so um I would say yeah sure if you have a test account already or you want to create it beforehand do it beforehand don't lose time with it in the hackathon if you don't have it just use your own unless you have privacy concerns But I don't think that um if you put a yeah if you put a link to the demo um account and fill it automatically so people will have access to your Gmail **[18:48]** address as long as you're okay with that um you can place that. >> Great. And if you're thinking about any of that like what um Deafany showed with the Brevo uh you're not able to connect to those type of email platforms with your personal. It has to be a domain that you own. Um, so something to just keep in mind there. >> Oh, actually, um, with me, I was I was able to connect with my own Gmail. So, the email [clears throat] itself goes out from my own Gmail. >> So, there you go. **[19:19]** >> Um, I I think so, it's like this scripted email address, you know, I think with my email, but it's essentially sending through Bravo, but my email's somewhere in there. >> Yeah. So, just so you know, >> um I'm going to take the qu this next question is, "Are there any features that may need payments while we build or will it be flexible?" So, I'll answer that one. Um don't spend money on anything during your 48 hours. Save it. Um you can set up anything. There's **[19:49]** usually like a free version out there or just set up like a demo. You don't have to like please don't spend money. [laughter] Um but here's one right back to Deafany. Um, what tools from Anastasia, what tools do you suggest for adding animations? Um, or does Lovable build animations? Um, Lovable actually does it for you? I think, um, it's a good question to ask to the guys from Get to Give because **[20:20]** they have really cool animations. I would I mean, you know, they are available um, they are in the alumni chat. So feel free to reach out to them and ask how did they do these cards animations but um to the extent that I have explored lovable does add whatever animations you want to add. So if you say overlap these three cards and then make them like flare out, I'm pretty sure it's going to be able to do that. >> Yeah. >> And how easy or difficult was it to **[20:51]** integrate API keys into the build? >> Oh, it's super easy. So, um, it even asks you to, um, put in the API key. So, if you say to Lovable, hey, I want to integrate with Brao, it's going to make it as easy for you as possible. It even gives you a link to Bravo to like sign in and to where to get your API key, even like direct deep link to wherever you need to go. So once you kind of discuss it and plan it with **[21:23]** lovable, I'm sure that you're just going to need to copy the API key and just put it in the chat and voila. >> Perfect. Um, let's see. Question from YouTube. I think I missed it, but how do you access the info from the weight list? Does it format format it as a CSV or a different type of file? H that's a good question. Um I think it's in your fi in the files of lovable **[21:54]** so you can see who fills it in. Just ask and again like it's something like Google it you know nowadays just ask lovable it's going to tell you exactly where that file is. Um because it might be keeping it in different files. I think um you know it the program also updates really um consistently like weekly they are showing you updates. So maybe there is um um there are some differences between each product or each **[22:24]** each um each build. So just ask lovable I would say. >> Great. And uh related to security how do you recommend to use lovable to protect the project from hackers? Yeah. Um, so Lovable, uh, I don't know, Marina, if you have the already seen or built something with Lovable, but once you get into the hackathon, security is going to be very much top of mind. Um, **[22:56]** and Whitney and Elena definitely will support you with all the questions in terms of the techniques of kind of overcoming security issues. And I'm sure since our time this has gotten a little bit better, but what happens is lovable flags each security issue to you already. So it detects what might be at risk. So a risky situation might be like a medium risk where you don't have to take even any action. It's just for your information. So you will know what are **[23:28]** the risks associated with the security breach or gap that you have in your product and what you should do. And you can automatically ask Lovable to fix it as well, which is just amazing and it's just the benefit of it. But definitely do I like what what I do is I say it in my PRD. I say I want um the external you know people the visitors who click my website to not access any information **[24:01]** put in by other uh visitors and these sort of specific prompts that you will put in your PRD will help lovable guide around and I think you can also again test these out before and do a little bit of digging into the channel that we are in with participants and alumni because this has happened like you know time repeats itself this has happened in season 2 a lot of us were stuck with **[24:32]** security issues so I suggest already typing in their security checking out the errors checking out the how to kind of overcome them >> absolutely and we do have I think two related micro webinars related to security this week. One of which is actually being hosted by one of the winners who had the site that you showed. So you everybody can ask about those card animations. >> Yeah, there you go. They're so helpful. Everyone is so helpful. Um ask us **[25:02]** anything in that chat. We are here for you and we're just enjoying the journey as as alongside you. And uh question from uh Tracy looks like if we're setting up something where we want to take user information, should we expect to bring in a third-party integration or can the lovable cloud keep that information for you? >> Mhm. Um I don't know about scaling it. Maybe Whitney knows a little bit more than I do about lovable cloud. As far as I **[25:33]** know, yes, Lovable Cloud is pretty great at doing just that. But uh Whitney if you if there's anything you want to add or Xavier go ahead. >> Yeah so uh lovable can hold uh levelable cloud can uh hold that for you. It is secure. It is just like super basease. Um and as you grow it can grow alongside you. Uh so that is absolutely possible. You do have the ability to use superbase directly if you want. It's whatever you are most comfortable with. Um **[26:04]** >> yeah about the last question I think uh I can also answer it with me you can too. It should be done in the 48 hours uh according to your specific product. So once you start building later in the game you're going to see the security issues show up. >> Absolutely. Um so you can start planning before Sunday but don't actually start building. So you'll want to address the security issues as you are building. Um, but wow, Deafany, thank you. Um, I **[26:35]** really love that you touched on so many different things because you could tell everybody's minds were right there being like, "Oh, I want to think about the design. I want to think about security." So, thank you. And I don't think it was a failure. I think that maybe that's just me and I don't want anyone to see themselves as a failure. >> Um, but the fact that you came back and are giving to this group is just so wonderful. Um, I know that you probably looked at season 1 and were like, "Wow, how kind of them." I hope you know that **[27:05]** you have season 3 now looking at you that same way. Um, so thank you so much. I love that you're in the Discord and we're all in this together. So, you're amazing. >> I am absolutely cheering for everyone. I'm sure you guys are going to just kill it. It's going to be amazing. and whatever you build like you know I think we don't give enough uh awards in this um hackathon everyone should everyone's a winner you know so definitely think about it this way I have built as you **[27:37]** saw like the women who built came out after this hackathon and it's just came out of my passion of she builds it inspired me to do something here locally um with my friends with other moms um you know how it is you know Last time when I was in season 2, I was also in the thick of it with my daughter who was 2 years old. So, I'm sure there are a lot of mothers who are building as they take care of their kids. So, definitely take a rest when you need, but also it's **[28:07]** going to be a really fun ride, the 48 hours. Um, just give it your all with the preparation as well. Believe in yourself, trust in yourself that it's going to go the way that you want it to go. And that's the most important thing. and good luck. Enjoy. >> Perfectly said. We're going to end with that. Thank you so so much and we'll see you in the chat.