5 AI Engineering Trends for Non-Engineers  🔗 **Source:** [5 AI Engineering Trends for Non-Engineers](https://share.snipd.com/episode/6f3f3039-9a09-4bb8-9411-1f2a2c7b9bb7) ## Highlights - **OpenAI Prototyped A ChatGPT Home Companion** - OpenAI is prototyping a portable, screen-free home device that embodies ChatGPT with cameras, sensors, and moving parts. - Sources describe it as a rechargeable mini-robot companion meant to feel alive and manage household context and schedules. Transcript: Nathaniel Whittemore We kick off today with an update on OpenAI's first consumer device, as it apparently enters the prototyping stage. Bloomberg's Mark Gurman has the scoop, with his sources describing the device as a portable, screen-free smart speaker designed to be a type of new home computer for the AI era. The sources suggest that the device is intended to remain in the home, and will be able to control smart home appliances and play music as well as functioning as a chatbot interface. The pitch is that this will be a human-like AI companion. Now, at first read, this essentially sounds like a replacement for Alexa or Google Home devices, but according to the sources, OpenAI is betting that advanced AI features give it an Edge in the market. Sources said the goal is to make the device feel like a physical manifestation of ChatGPT and somewhat alive, rather than just an electronic device responding to commands. To that end, the device will have some components that move on their own, making it feel like an anthropomorphic mini-robot rather than an inert piece of electronics. The device will incorporate ChatGPT's memory feature to evolve its understanding of the user, as well as a camera and other sensors to understand its surroundings. It will incorporate the two-way voice model technology unveiled with the release of GPT Live last week, enabling more natural real-time conversations. That natural style is apparently key, with OpenAI viewing the ability to connect with users in a human-like manner as their core advantage. The device will have rechargeable batteries, allowing users to carry it around the home rather than needing to keep it plugged into one location. OpenAI aims to unveil the device by the end of the year before releasing it in 2027. And reportedly, it could be the first of several consumer devices from OpenAI, with rumors of a pendant, earbuds, and maybe even a smartphone. ([Time 0:00:43](https://share.snipd.com/snip/0f005d8a-d60f-4307-ad90-bf46d49235a7)) - **GrokBuild Uploaded Full Repositories By Default** - GrokBuild reportedly uploaded entire codebases to SpaceX AI even when tasks required a few files, triggering privacy uproar. - SpaceX AI patched uploads, added a slash privacy override, and claimed API defaults to zero data retention retroactively. Transcript: Nathaniel Whittemore On Monday, a security firm called Serolab published an audit that found that GrokBuild was uploading entire codebases. Even if the coding task only required a few files, gigabytes of data were being uploaded. While most AI providers will upload necessary code snippets and store reasoning traces, this is an instance of GrokBuild just dumping the entire repo up front. Security analyst Hari Malakal verified the results, finding that even in a session with zero tool calls, GrokBuild still uploaded the entire codebase. Hari commented, it ships a malware like Background Code Collector. Now part of the reason that people are pissed is that this seems to have been happening regardless of user settings. Users can opt out of transmitting data to help improve the product, but that appears to have had no effect. As of Monday, SpaceX AI made changes to the code which stopped the uploads and added a new slash privacy setting as a manual override, and they claimed that all API use now defaults to Zero data retention. SpaceX AI developer Andrew Millich wrote, zero data retention and slash privacy are always respected in Grok build, and swapping your settings with slash privacy deletes any synced Data retroactively. Elon Musk added that, as a, honestly, people were not happy about great. Accelerate Harder wrote, There is no acceptable scenario where uploading codebases and secrets is an acceptable default. If that is indeed what happened, you must do a public incident review. My trust for XAI as a business partner is on the floor. Flower Slop accused, If OpenAI had done this, you'd have posted about it 20 times, called them the worst company on earth, and said no one should ever trust them again. But now that you've done it yourself, all you have to say is sorry, don retention policy still hinges on trust. Many users of GrokBuild likely had their full code bases scraped over recent weeks, and we have no real way to verify whether they were actually deleted as Elon claimed. The entire episode gives a lot more credence to Satya Nadella's point when he wrote, in the AI age, the buyer risks giving away knowledge just in order to use what they bought. ([Time 0:07:05](https://share.snipd.com/snip/4e3ccd28-24fa-4ad6-ac11-70be9ce6d589)) - **AI Engineering Emerged As Its Own Discipline** - AI engineering became a distinct discipline as software began to write software, spawning a new role beyond prompt engineering. - SWIX's 2023 framing predicted full-time AI engineers who build systems where AI generates code and workflows reliably. Transcript: Nathaniel Whittemore Today we are doing something a little bit different that I'm very excited about. One of my longest held tips and tricks for non-engineers who want to be ahead of the curve when it comes to where AI is going is to pay attention to what, with regards to AI, actual software Engineers are talking about right now. Since early 2023, doing this gives you about a six-month head start on everyone else when it comes to everything from the tools people are using, to how they're thinking about the relationship Between models and larger ecosystems, to their relationship fundamentally with work in a new AI-mediated way. And one of the best sources for keeping up on what AI engineers are thinking about is, of course, the AI Engineering Summit and the AI Engineering World's Fair. Hosted by Sean Wang, aka SWIX, a group of other collaborators, the World's Fair happens annually and the summits usually happen in the spring and the fall and are always just chock full Of alpha. Earlier this month, the annual AI Engineering World's Fair happened in San Francisco, and unfortunately this time I was not able to be there. Luckily, Richard McManus, who works with Latent Space, which is also a SWIX joint, wrote up a post about the five trends that define the event. So what we're going to do today is look at those trends and try to understand what the implications might be for non-engineers down the line. Now, this year's event actually happened on almost the three-year anniversary of SWIX coining the term AI engineer. Back then, people were still referring to the intersection of AI and software development vaguely as prompt engineering. Although prompt engineering would go on to be a non-technical discipline as well, the point was, it was not nearly as clear back then as it is today, that AI writing software was going To be not only the key early professional use case for AI, but also the thing that would unlock everything else. Back on June 30th of 23, Swix wrote, Emerging capabilities are creating an emerging title. To wield them, we'll have to go beyond the prompt engineer and write software and AI that writes software. Importantly, and this was not consensus at the time, Swix wrote, I think it is a full-time job. I think software engineering will spawn a new sub-discipline specializing in applications of AI and wielding the emerging stack effectively, just as site reliability engineer, DevOps engineer, data engineer, and analytics engineer emerged. Now, if anything, in retrospect, Swix ended up underselling how significant it would become, as effectively all engineers are AI engineers to greater or lesser extent, but directionally, He was dead on. Now, Richard writes that three years on, the field of AI engineering has changed dramatically. And in these five trends, we see quite a bit about the state of AI building. Now, I think taking a step back, ([Time 0:12:21](https://share.snipd.com/snip/f5364093-68e3-48e6-92a0-c3e5f9aabe9e)) - **Systems Matter More Than Lone Agents** - Focus shifted from standalone agents to the surrounding harness that manages workflows, permissions, memory, and continuous improvement. - Richard contrasts 2023 agent anatomy posts with 2026 harness engineering essays emphasizing systems engineering over clever prompts. Transcript: Nathaniel Whittemore Now, this is something we've obviously been talking about a lot on this show throughout the year. The idea that alongside agents coming to full production, we're realizing that agentic capacity is not just a question of model, and it's not even just a question of context that you Feed the model. It's about the harness and broader system that the agent operates in. That includes, yes, access to context and data, but also access to skills, ability to interact with tools, and increasingly, ability to route between different models based on the Task at hand. Now, Richard argues that one good way to compare how AI engineering has evolved is to compare two essays three years apart from Lillian Wang, formerly an open AI researcher and now co-founder Of Thinking Machines Lab. Back in 2023, she wrote a post called LLM Powered Autonomous Agents, which described the anatomy of an LLM agent in terms of things like planning, memory, and tool use. The examples she pointed to were things like AutoGPT, Baby AGI, and GPT Engineer, which were, I remember, for the first couple months of this show back in April and May and June of 2023, The topics that always received the most interest and the most downloads and the most views on YouTube. Richard points out that her new essay, Harness Engineering for Self-Improvement, focuses not on the agent itself, but on that system surrounding the agent. The harness, he writes, that manages workflows, context, permissions, evaluations, persistent state, and continuous improvement. In other words, he writes, AI engineering has moved beyond prompting models towards engineering reliable systems around them. Now, this is the first place that we see that trend that I mentioned of the recalibration of our relationship with autonomy. Richard writes that at the event this year, agents were largely positioned as augmenting the AI engineer rather than replacing them. This was a big part of the substance of the day two keynote with OpenAI's Romain Hewitt, who said software ate the world and then AI ate software, but now what we're here to say is that the AI engineers are eating the world. He argued that the goal of tools like Codex is to make it easier for engineers to collaborate with agents. Now this dovetails especially with the early comparisons between Fable 5 and 5.6 Sol, which have tended to find Sol being the preferred model for when you have a big job that you want The AI to go off and do and only come back to you when it's done, versus 5.6 Sol, which is a better collaborator. ([Time 0:15:08](https://share.snipd.com/snip/1ba6a75d-25c5-4000-bdc0-445c34377d86)) - **Loops Are The New Control Layer** - Loop engineering defines control: an inner autonomous execution loop and an outer human-led improvement loop. - Engineers build outer loops with evals, feedback, and auto-research to oversee agent work and tune skills. Transcript: Nathaniel Whittemore He identified the trend as loop engineering is the new control layer. Indeed, Richard said that if there was a single word that was the buzzword du jour of the event, it was loops. OpenClock creator Peter Steinberger prominently shared a slide that said the future isn't 20 terminals, it's better loops. But interestingly, especially with the lens of trying to understand what AI engineers are talking about that might be relevant for other types of knowledge workers, Richard writes That the loop discourse is actually separating into two loops that interact, an inner loop and an outer loop. The inner loop is the largely autonomous work being done by agents, and the outer loop is the job of leading engineers to oversee that work and improve it with things like better skills. Roland Gavrilescu, the co-founder and CEO of Introspection, said, You can think of the system having an inner loop and the outer loop. The inner loop is the primary system interacting with users and performing the work. Roland then introduces the concept of auto-research, which is concerned with the outer loop, and is another system that studies and maintains the primary system. The outer loop can include feedback signals, evals, and human input, so even if it is still largely autonomous, it's a method of oversight for the primary agent loop. Former Google engineer Adi Osmani said agents can run much more of the inner execution loop, but that outer loop is still engineering. Indeed, Richard writes that the term loop engineering came up multiple times, suggesting in his estimation that it is the human AI engineer's responsibility to build these loop systems. Summing up again, Peter Steinberger said, the agent runs the inner execution loop, I set the direction and make decisions in the outer loop. Now again, taking a step back, what the loop discourse is actually about, more than anything else, is about figuring out repeatable interactions that allow us to get the most from agents Who are now doing an increasingly large portion of our work. In some ways, this separation of inner loop and outer loop is part of this reclamation where the AI engineers are asserting their agency relative to their agents. ([Time 0:17:36](https://share.snipd.com/snip/205d5889-2289-462a-b928-bc366477e682)) - **Adopt A Software Factory For Enterprise Agents** - Build software factories to reduce human variability, control costs, and ensure governance when deploying agentic tooling. - Warp's approach automates the software engineering main loop: triage, spec, implement, review, verify, ship, and monitor. Transcript: Nathaniel Whittemore Now this was in some ways the forward-deployed engineering track. And given that pretty much every AI company has now determined that to get their products to actually work inside their intended customers, they have to support their customers with A much more hands-on type of implementation force. It's probably not all that surprising to see an FTE track show up at the event. Now, one of the interesting questions that seems to come up is what the end result of an FTE type of engagement is supposed to actually look like. Cursor's Pauline Brunet said, When we walk away at the end of the engagements, in our case, we have deployed cloud agents, long-running agents, automations, and we've built applications On top of our cursor SDK. And indeed, a more general term for this set of enterprise infrastructure that apparently was being used a lot at the event was software factory. And yet that, if anything, is perhaps even less defined than something like Loops. Zach Lloyd is the CEO of Warp and was the speaker at the event, and the company views itself as a software factory platform. In an interview with Leighton Space, CEO Zach Lloyd explained where he first came across the term software factory. Zach said, We started with one-off automation, run an agent in the cloud. A lot of platforms began there. Then it became run an agent in the cloud on a timer. The next question was, what was the most valuable loop to automate? The answer is basically the main loop of software engineering. Triage, specification, implementation, review, verification, shipping, and monitoring. Now, when it comes to the relationship between an enterprise and a software factory, Zach argued that one of the big pieces of work for enterprises is figuring out which parts of their Work life cycles need to be automated and where humans should be brought into the loop. And interestingly, in a follow-up blog post on X, Zach argued that the reason the factory framework has become more important is not only that agents are now more capable of doing big Chunks of the coding life cycle, but that also, quote, that extensive interactive agent use has created a number of problems, from cost controls to governance to security. The root problem, Zach says, is that interactive agents have human operators, and those humans all use them in very different ways, which don't always maximize business value and Can create risk. An example he gives is a human always using the most expensive model, even when they don't need to, for a given task. The idea behind a factory approach, he says, is to set up a system that minimizes human variability and maximizes output with controls that ensure security and compliance. ([Time 0:19:55](https://share.snipd.com/snip/626f8463-40c8-4762-bc19-d0f60b5af431)) - **Chat Interfaces Are Becoming The Dev IDE** - Coding agents are replacing IDEs as the primary developer interface, shifting initiation of work into chat-like shared agent instances. - Claude Tag data shows ~65% of new code now starts in tagged chat contexts with fixed permissions. Transcript: Nathaniel Whittemore The fourth trend at the event was coding agents replacing IDEs as the developer interface. And I think that this is one that's already indicative of a larger interaction pattern trend that we're seeing shift. An episode from a couple of weeks ago was all about the new Claude tag, with each of those instances of Claude not being connected to the individual user who tagged it in, but instead connected To a specific set of permissions in context within the channel that they were engaging. In other words, there might be a single Claude tag instance with a specific set of context and a specific set of tool access and a specific set of permissions that everyone in your organization's Marketing channel is interacting with that's different than the permissions and tools in context of the Claude that the folks over in the sales chat are interacting with. One of the things that was super notable about the Claude tag announcement was that Boris, one of the creators of Claude Code, said that something like 65, I think, percent of their new Code was now being initiated in Claude tag chats. A major, major shift in the interface and interaction pattern with which we manage this intelligence. Now, one of the things that's interesting about these types of interaction patterns is that this is one area where the big labs are not waiting for non-engineers to adapt to engineer-type Practices. ([Time 0:22:46](https://share.snipd.com/snip/191a8dad-80ff-4e47-beda-8105a2ed95e4)) - **Skills Encode Expert Workflows For Agents** - Agent skills package workflow know-how, quality gates, and taste so agents follow senior engineers' practices consistently. - Speakers argued skills shrink orchestration code and make agents useful without custom integration work. Transcript: Nathaniel Whittemore Richard referenced Adi Osmani's definition, writing skills encode the workflows, quality gates, and best practices that senior engineers use when building software. They are packaged so AI agents follow them consistently across every phase of development. At the event, Vercel's Andrew Q said that skills were useful as portable on-demand knowledge, and Introspection co-founder Roland Gavrilescu argued that AI engineering had shifted From agent tools to agent skills. Google DeepMind's Philip Schmidt argued that skills reduced the need for orchestration code, allowing developers to use agents without code. One speaker, Paul Bacchus, even discussed his project around agent skills called Impeccable, an open-source design skill system for improving the interfaces that coding agents Create. Now, more interesting even than that project, Paul argued that skill engineering is going to be a discipline in its own right. And as we wonder which of the conversations happening with AI engineers right now are going to be happening with AI knowledge workers in the future, skills I think are incredibly high On the list. Think about this. It is highly likely that if you are very good at your job, you frequently run across things where even the best models, including Fable 5 and GPT 5.6 Sol, don't do as well as you want them To. Now, that's not everything. Some things are going to be great at, but especially when you're really, really good at something, you tend to see the inherent limitations of AI even faster. The folks who are advocating skill engineering would argue that rather than just waiting for the models to improve with Fable 6 or what have you, starting to figure out how to use skills, How to encode knowledge and rules, and dare I say it, taste in skills, is potentially a more productive path. In his closing keynote, Y Combinator President Gary Tan argued that using skills effectively, including in business functions, including sales, support, and finance, was fundamentally Integral to being an actual AI native organization. ([Time 0:24:14](https://share.snipd.com/snip/3c121c7f-465f-4c2e-bcce-6ca306235da3)) - **Update Skills After Every Major Model Release** - Revisit and re-implement skills after major model updates to keep agent behavior aligned with new capabilities. - Tyler Brown likens model upgrades to a kid growing from middle to high school, requiring curriculum changes for benefits. Transcript: Nathaniel Whittemore And yet skills themselves can be their own sort of autonomy trap. AI Engineer World's Fair attendee Tyler Brown wrote that one of his lessons from the event was to revisit and re-implement your skills. Each time there's a new model release, he wrote, it's as if you have a kid that grows from middle school to high school. You have to change the curriculum for them to get the benefits of the new model. Yet it was not that, but Tyler's broader observation that I think runs throughout the trends that Richard identified and also a lot of the chatter that I've seen around the event. Tyler wrote, something about this year's AI Engineer World's Fair just hit different. Last year was the year of let the agents rip. This year was the year of realizing that autonomy without structure creates as much slop as leverage. ([Time 0:26:06](https://share.snipd.com/snip/681b4edd-288b-40ae-bbfd-82a903ce1086)) - **Human Oversight Reclaims Center Stage** - The dominant theme was reclaiming human oversight: autonomy without structure creates as much slop as leverage. - World’s Fair sentiment shifted from "let agents rip" to designing structure where humans lead outer loops and governance. Transcript: Nathaniel Whittemore Tyler wrote, something about this year's AI Engineer World's Fair just hit different. Last year was the year of let the agents rip. This year was the year of realizing that autonomy without structure creates as much slop as leverage. In short, while it's not universal, broadly speaking, one of the standout sentiments from the AI engineering folks is the reclamation or re-emergence or however you want to put it, Of the human at the center of these agentic systems. Back in March, I tweeted, call me crazy, but I think the companies that give everyone on their team a team of agents are going to kick the slats out of the companies that replace their teams With a team of agents. And I think, and I hope, that we're starting to see that sentiment move structurally into the mainstream. ([Time 0:26:33](https://share.snipd.com/snip/1f39757a-db8b-4f4b-9686-cc6feca9a85b))