Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

LLMs work best when the user defines their acceptance criteria first

blog.katanaquant.com

Post reply on HN
Print view View original

181–190 of 460 posts

Re: LLMs work best when the user defines their acceptance criteria first

#181

Post by consumer451 »

dragonwriter wrote: post #179
consumer451 wrote: post #172

Nitpick/question: the "LLM" is what you get via raw API call, correct? If you are using an LLM via a harness like claude.ai, chatgpt.com, Claude Code, Windsurf, Cursor, Excel Claude plug-in, etc... then you are not using an LLM, you are using something more, correct? An example I keep hearing is "LLMs have no memory/understanding of time so ___" - but, agents have various levels of memory. I keep trying to explain th…

> Nit pick/question: The LLM is what you get via raw API call, correct? You always need a harness of some kind to interact with an LLM. Normal web APIs (especially for hosted commercial systems) wrapped around LLMs are non-minimal harnesses, that have built in tools, interpretation of tool calls, application of what is exposed in local toolchains as “prompt templates” to transform the context structure in the API cal…

In meetings, I try to explain the roles of system prompts, agentic loops, tool calls, etc in the products I create, to the stakeholders.

However, they just look at the whole thing as "the LLM," which carries specific baggage. If we could all spread the knowledge of what is actually going on to the wider public, it would make my meetings easier, and prevent many very smart folks who are not practitioners from saying inaccurate stuff.

Re: LLMs work best when the user defines their acceptance criteria first

#182

Post by ashdksnndck »

bryanrasmussen wrote: post #62
pornel wrote: post #36

Their default solution is to keep digging. It has a compounding effect of generating more and more code. If they implement something with a not-so-great approach, they'll keep adding workarounds or redundant code every time they run into limitations later. If you tell them the code is slow, they'll try to add optimized fast paths (more code), specialized routines (more code), custom data structures (even more code).…

maybe there should be an LLM trained on a corpus of a deletions and cleanup of code.

I think this is in the training data since they use commit data from repos, but I imagine code deletions are rarer than they should be in the real data as well.

Re: LLMs work best when the user defines their acceptance criteria first

#183

Post by movedx01 »

krackers wrote: post #89
bryanrasmussen wrote: post #62

Earlier quoted context omitted.

maybe there should be an LLM trained on a corpus of a deletions and cleanup of code.

I'm guessing there's a very strong prior to "just keep generating more tokens" as opposed to deleting code that needs to be overcome. Maybe this is done already but since every git project comes with its own history, you could take a notable open-source project (like LLVM) and then do RL training against against each individual patch committed.

Perhaps the problem is that you RL on one patch a time, failing to capture the overarching long term theme, an architecture change being introduced gradually over many months, that exists in the maintainer’s mental model but not really explicitly in diffs.

Re: LLMs work best when the user defines their acceptance criteria first

#184

Post by staplers »

consumer451 wrote: post #181
dragonwriter wrote: post #179

Earlier quoted context omitted.

> Nit pick/question: The LLM is what you get via raw API call, correct? You always need a harness of some kind to interact with an LLM. Normal web APIs (especially for hosted commercial systems) wrapped around LLMs are non-minimal harnesses, that have built in tools, interpretation of tool calls, application of what is exposed in local toolchains as “prompt templates” to transform the context structure in the API cal…

In meetings, I try to explain the roles of system prompts, agentic loops, tool calls, etc in the products I create, to the stakeholders. However, they just look at the whole thing as "the LLM," which carries specific baggage. If we could all spread the knowledge of what is actually going on to the wider public, it would make my meetings easier, and prevent many very smart folks who are not practitioners from saying i…

  If we could all spread the knowledge of what is actually going on to the wider public, it would make my meetings easier, and prevent very smart folks from outside the field from saying dumb-sounding stuff.
This is an example of why LLMs won't displace engineers as severely as many think. There are very old solved processes and hyper-efficient ways of building things in the real world that still require a level of understanding many simply don't care or want to achieve.

Re: LLMs work best when the user defines their acceptance criteria first

#185

Post by lukan »

Jweb_Guru wrote: post #121
Foobar8568 wrote: post #114

Earlier quoted context omitted.

That's the reality nobody really wants to say.

It's not reality. I'm really not a fan of the way that people excuse the really terrible code LLMs write by claiming that people write code just as bad. Even if that were true, it is not true that when you ask those people to do otherwise they simply pretend to have done it and forget you asked later.

"Even if that were true, it is not true that when you ask those people to do otherwise they simply pretend to have done it and forget you asked later."

I admire your experience with people.

Re: LLMs work best when the user defines their acceptance criteria first

#186

Post by abm53 »

iLoveOncall wrote: post #154
unlikelytomato wrote: post #70

Earlier quoted context omitted.

This is why I'm confused when people say it isn't ready to replace most of the programmer workforce.

Really? Because this perfectly explains why it will never replace them: it needs an exact language listing everything required to function as you expect it. You need code to get it to generate proper code.

I think GP was a joke about the ability of a typical programmer.

I certainly read it as one and found it funny.

Re: LLMs work best when the user defines their acceptance criteria first

#187

Post by alexhans »

> The vibes are not enough. Define what correct means. Then measure.

Pretty much. I've been advocating this for a while. For automation you need intent, and for comparison you need measurement. Blast radius/risk profile is also important to understand how much you need to cover upfront.

The Author mentions evaluations, which in this context are often called AI evals [1] and one thing I'd love to see is those evals become a common language of actually provable user stories instead of there being a disconnect between different types of roles, e.g. a scientist, a business guy and a software developer.

The more we can speak a common language and easily write and maintain these no matter which background we have, the easier it'll be to collaborate and empower people and to move fast without losing control.

- [1] https://ai-evals.io/ (or the practical repo: https://github.com/Alexhans/eval-ception )

Re: LLMs work best when the user defines their acceptance criteria first

#188

Post by mirsadm »

leke wrote: post #138
pornel wrote: post #36

Their default solution is to keep digging. It has a compounding effect of generating more and more code. If they implement something with a not-so-great approach, they'll keep adding workarounds or redundant code every time they run into limitations later. If you tell them the code is slow, they'll try to add optimized fast paths (more code), specialized routines (more code), custom data structures (even more code).…

i wonder if the solution is to just ask it to refactor its code once it's working.

I do this all the time but then you end up with really over engineered code that has way more issues than before. Then you're back to prompting to fix a bunch of issues. If you didn't write the initial code sometimes it's difficult to know the best way to refactor it. The answer people will say is to prompt it to give you ideas. Well then you're back to it generating more and more code and every time it does a refactor it introduces more issues. These issues aren't obvious though. They're really hard to spot.

Re: LLMs work best when the user defines their acceptance criteria first

#190

Post by mirsadm »

88j88 wrote: post #98

100% I found that you think you are smarter than the LLM and knowing what you want, but this is not the case. Give the LLM some leeway to come up with solution based on what you are looking to achieve- give requirements, but don't ask it to produce the solution that you would have because then the response is forced and it is lower quality.

100% dependent on the person driving it
Post reply on HN