[kiitos]

> My prompts specify very precisely what should be implemented. I specified the public API and high-level design upfront. I let the AI come up with its own storage schema initially but then I prompted it very specifically through several improvements (e.g. "denormalize this table into this other table to eliminate a lookup"). I designed the end-to-end encryption scheme and told it in detail how to implement it. I pointed out bugs and explained how to fix them. And so on.

OK. Replace "[expected] deterministic output" with whatever term best fits what this block of text is describing, as that's what I'm talking about. The claim is that a sufficiently-precisely-specified prompt can produce reliably-correct code. Which is just clearly not the case, as of today.

[tptacek]

I don't even think anybody expects reliably-correct code. They expect code that can be made as reliably as they themselves could make code, with some minimal amount of effort. Which clearly is the case.

[kiitos]

Forget about reliably-correct. The code that any current-gen LLM generates, no matter how precise the prompt it's given, is never even close to the quality standards expected of any senior-level engineer, in any organization I've been a part of, at any point in my career. They very much never produce code that is as good as what I can create. If the LLM-generated code you're seeing passes this level of muster, in your view, then that's really a reflection on your situation(s), and 100% not any kind of truth that you can claim as part of a blog post or whatever...

[kentonv]

> The code that any current-gen LLM generates, no matter how precise the prompt it's given, is never even close to the quality standards expected of any senior-level engineer, in any organization I've been a part of, at any point in my career.

You are just making assertions here with no evidence.

If you prompt the LLM for code, and then you review the code, identify specific problems, and direct the LLM to fix those problems, and repeat, you can, in fact, end up with production-ready code -- in less time than it would take to write by hand.

Proof: My project. I did this. It worked. It's in production.

It seems like you believe this code is not production-ready because it was produced using an LLM which, you believe, cannot produce production-ready code. This is a cyclic argument.

[kiitos]

> If you prompt the LLM for code, and then you review the code, identify specific problems, and direct the LLM to fix those problems, and repeat, you can, in fact, end up with production-ready code

I guess I will concede that this is possible, yes. I've never seen it happen, myself, but it could be the case, at some point, in the future.

> in less time than it would take to write by hand.

This is my point of contention. The process you've described takes ages longer than however much time it would take a competent senior-level engineer to just type the code from first principles. No meaningful project has ever been bottle-necked on how long it takes to type characters into editors.

All of that aside, the claim you're making here is that, speaking as a senior IC, the code that an LLM produces, guided by your prompt inputs, is more or less equivalent to any code that you could produce yourself, even controlling for time spent. Which just doesn't match any of my experiences with any current-gen LLM or agent or workflow or whatever. If your universe is all about glue code, where typing is enemy no. 1, and details don't matter, then fair enough, but please understand that this is not usually the domain of senior-level engineers.

[simonw]

"the code that an LLM produces, guided by your prompt inputs, is more or less equivalent to any code that you could produce yourself, even controlling for time spent"

That's been my personal experience over the past 1.5 years. LLMs, prompted and guided by me, write code that I would be proud to produce without them.

[kentonv]

I have only claimed that for this particular project it worked really well, and was much faster than writing by hand. This particular project was arguably a best-case scenario: a greenfield project implementing a well-known standard against a well-specified design.

I have tried using AI to make changes to the Cloudflare Workers Runtime -- my usual main project, which I started, and know like the back of my hand, and which incidentally handles over a trillion web requests every day -- and in general in that case I haven't found it saved me much time. (Though I've been a bit surprised by the fact that it can find its way around the code at all, it's a pretty complicated C++ codebase.)

It really depends on the use case.

[59nadir]

It's possible kiitos has (or had?) a higher standard in mind for what should constitute a senior/"lead engineer" at Cloudflare and how much they should be constrained by typing as part of implementation.

Out of interest: How much did the entire process take and how much would you estimate it to take without the LLM in the loop?

[kentonv]

> It's possible kiitos has (or had?) a higher standard in mind for what should constitute a senior/"lead engineer" at Cloudflare and how much they should be constrained by typing as part of implementation.

See again here, you're implying that I or my code is disappointing somehow, but with no explanation for how except that it was LLM-assisted. I assert that the code is basically as good as if I'd written it by hand, and if you think I'm just not a competent engineer, like, feel free to Google me.

It's not the typing itself that constrains, it's the detailed but non-essential decision-making. Every line of code requires making several decisions, like naming variables, deciding basic structure, etc. Many of these fine-grained decisions are obvious or don't matter, but it's still mentally taxing, which is why nobody can write code as fast as they can type even when the code is straightforward. LLMs can basically fill in a bunch of those details for you, and reviewing the decisions -- especially the fine-grained ones that don't matter -- is a lot faster than making them.

> How much did the entire process take and how much would you estimate it to take without the LLM in the loop?

I spent about five days mostly focused on prompting the LLM (although I always have many things interrupting me throughout the day, so I wasn't 100% focused). I estimate it would have taken me 2x-5x as long to do by hand, but it's of course hard to say for sure.

[59nadir]

> See again here, you're implying that I or my code is disappointing somehow, but with no explanation for how except that it was LLM-assisted. I assert that the code is basically as good as if I'd written it by hand, and if you think I'm just not a competent engineer, like, feel free to Google me.

I think you're reading a bit too deeply into what I wrote; I explained what I interpreted kiitos' posts as essentially saying. I realize that you've probably had to deal with a lot of people being skeptical to the point of "throwing shade", as it were, so I understand the defensive posture. I am skeptical, but the reason I'm asking questions (alongside the previous bit) is because I'm actually curious about your experiment.

> It's not the typing itself that constrains, it's the detailed but non-essential decision-making. Every line of code requires making several decisions, like naming variables, deciding basic structure, etc. Many of these fine-grained decisions are obvious or don't matter, but it's still mentally taxing, which is why nobody can write code as fast as they can type even when the code is straightforward. LLMs can basically fill in a bunch of those details for you, and reviewing the decisions -- especially the fine-grained ones that don't matter -- is a lot faster than making them.

In your estimation, what is your mental code coverage of the code you ended up with? Do you feel like you have a complete mapping of it, i.e. you could get an external request for change and map it quickly to where it needs to be made and why exactly there?

[bdangubic]

100% this. I have same proof… In productions… across 30+ services… hourly…

[tptacek]

The genetic fallacy is a hell of a drug.