I have a hard enough time explaining why “yet another bespoke application on some unmanaged azure resources” is a bad idea when they have more reasonable alternatives at hand.
Now these goofs can (very nearly) press a button and do it anyway, with no comprehension of the consequences. It’s high fives and pats on the back, until I’m cleaning up the mess.
Knowing what not to do has always been important, but it is perhaps more important than ever.
I have a design principles + tech preferences doc I force llms "lint" their approach against. It's not perfect but it helps. I call it a bias field, pushes them toward hopefully the happy and harmonious (with the rest of the system) paths. Obviously this is only partial and imperfect enforcement, but if it's applied to everything consistently it naturally encodes some self-consistency and harmony.
I’ve done the same but it’s a moving target as models advance and I find half of my points are ignored until I’m prompting “No wtf why are you still trying to symlink the global Python executable just use the virtual environment that’s already activated”.
Anyway, companies are pouring billions into improving AI tooling user experience so most of what I do manually I just anticipate to be a waste of time. There’s no way my hobby fiddling will outpace whatever gets released in the next couple months.
In the meantime, real linting does work pretty well, if you can write a detector for whatever antipattern you find LLMs fall into (like multiline comments).
Although, given opus 4.8->5.0 experience within claude code, I don't know if I'd agree they necessarily improve things with time. But that's just my experience. I should probably read their release notes, maybe just a tweak to my claude.md would help
I get patches into the biggest AI projects all the time to make inference faster on my 3090 and all you have to do is read the contribution docs and open a small PR.
It’s not a cabal of super geniuses. We are literally in the Wright Brothers era of AI.
You can trivially outpace what big companies are doing it’s insane, why do you think so many innovations are coming out of scrappy Chinese labs? They are willing to engage instead of being defeatist about it.
Not my core competence and I’m not interested in making it a core competence while the field is still rapidly changing. I find harness engineering interesting, but not lucrative enough to push one of the many very important other things on my plate to the backburner.
I don’t think Python language developers are super geniuses either, and I’ve never submitted code for consideration.
Remove the dev from the equation. Suppose you need to produce circumstances for the slop machine to succeed on its own (imagine a loop).
What is missing? A good enough description of the current behavior, desired changes and how it should work? Understanding of the business domain or the system in question, its architectural choices? Enough guardrails and linters and tests to ensure that the solution will indeed be testable and work end to end? Cause a lot of the things that will help human devs will also help LLMs make their output more workable - for example ADRs in the repo and custom in-house tools to ensure they follow architectural conventions and practices (or just something like ArchUnit). Realistically, how close to a finished implementation must you get and how far backwards from that can you go before its output degrades too much?
I tried figuring that out myself on a personal project where I mostly want what I want in GitHub issues and then feed them to Kepler, which spawns OpenCode in worktrees and also gives me a UI to follow the execution. Most models, even SOTA, occasionally implement things wrong, but it's suprising how far you can take things when you've spent a dozen hours planning out what you want the end result to be vs not to be (talking about the goals for an entire system, maybe an hour of planning for a more bounded type of task).
Or, slightly changing this. The source text needs to speak the correct vocabulary and language to produce a good completion. See the chat where Terry Tao is doing maths with an LLM. There’s _no way in hell_ I could get to his output because I just have no idea, and can’t speak the language.
Same with any field.
The edge of correct/better when facing ambiguity is very fuzzy, all models from the past 6 month or so have similar random ways of spinning between too-literal avenues and oddly misplaced misled fixations. Taking the right initiatives in face of uncertainty is definitely AGI, and its not there, and perceptrons + attention layers just ain't got what it takes no matter how hard you push.