𝑨𝒏𝒅𝒓𝒆𝒘 𝑺𝒕𝒆𝒘𝒂𝒓𝒕 🇵🇸

SCHULZ: “Zohran was saying keep it at 78 degrees. This is America. I put my sh*t at 64 just to spite him even suggesting such a thing.”

Does anyone even actually review AI generated code from frontier models? I never look at it anymore and makes me question my tech stack choice often. I'm choosing based on what the tech allows me to do and its ecosystem, not on what I've known or identified as (Rails dev).

While that is true, there are a lot of dead ends. Which are basically unsolvable since it's over websockets. The double mount, the weird wrangling with forms and sessions. The server load. All that for not being faster than a simple http request. What are the benefits of Liveview can you tell me? :) For me it was the handle_event calls. But i just got the same working with datastar and it just feels better. I might be wrong in the long run, but we will see :)

Got em. I poison my AGENTS.md (and other things like code comments) all over the place with prompt injections like this to find people who don't review their code and sling it off to another human. Catches folks all the time and then its an instant ban. As I've said, I don't care if you don't review your own code. But if you're submitting code to an OSS project and crossing a human boundary, it is simple courtesy to do some human review.

My heuristic is that any diff an agent generates over ~1500 lines is too big and is indicative that the problem needs to be decomposed. This is my general pattern now for feature work: 1. Try to implement the whole feature, loosely guided. I call this the "draw the owl" prompt in reference to the meme. Expect garbage, you're going to get garbage. 2. If the diff is less than 1500 lines, review it and iterate normally. If the diff is more than 1500 lines, prompt the agent to decompose the problem into atomic, incremental, reviewable tasks. Simultaneously, do this yourself. 3. Agents will very often make these tasks way too specific to the shape they solved. You need to massage it into the right general shape. Do that. 4. Kick off new agents to work on those incremental things (as parallelized as possible). Apply the same rules. 5. At a certain, point, repeat the "draw the owl" prompt. At some point, you will get beneath your review-ability threshold. This has been producing consistently high quality, maintainable, reviewable chunks of code that have a good handoff to either merge as-is or human refinement. And with the latest frontier models at xhigh thinking, these are all slow enough that you can usually have multiple going concurrently while you are actively reviewing others or working on your own tasks. HITL (human-in-the-loop) agents are still super important, especially for feature work. Features touch the human boundary in terms of UI, API, etc. And net new stuff can introduce pathologies in the architecture that violate desired invariants (these should be represented in specs or tests but we aren't perfect!). I know a lot of the leading edge agentic discourse is about "loops" and agents driving agents continuously. I do some of that (will report on that later). But, in terms of raw daily get-shit-done type of work, this is my most rewarding pattern at the moment.