Understanding is the new bottleneck • Hacker News Board

Re: Understanding is the new bottleneck

#21

Post by cyanydeez » Thu, Aug 13, 2026, 8:23 PM UTC

w10-1 wrote: post #19

I agree with the problem but not the solutions. The problem pre-dates LLM's: writing code that "works" but breaks the underlying model. Because it works, it always sounds reasonable and doesn't raise any flags. Only someone - human or LLM - who holds the model as the standard would see that this working solution breaks the model. (In theory, the model is to preserve scaling, flexibility or some other systemic feature…

understanding a different modality of model interaction gave me proper insight into the specific problem. In visual models, even if the model understands the concept of face, or hand, or whatever, it doesn't know how to de-dupe a statement like "count the number of faces" until you give it a countable reference frame, so it can internally, place a box around a face and give that a coordinate, and then it can collect all the coordinates, and suddenly it's counting face in a picture.

The same thing happens in code. Things we're happily shifting from context to context, the model itself isn't doing. When it reads file1 for the main() clause, it will easily read file2's main() clause as the same. It'll internally merge these.

So if you do want to work with these models to achieve complex tasks, you basically do have to go reverse centaur and bend the code base to it's blindness. You can't use the same function names across the code base; each one needs to be dstinguishable; same thing with variables that represent seperate entity relationships.

You do that, and it suddenly because a whole lot smarter.

Re: Understanding is the new bottleneck

#22

Post by avaer » Thu, Aug 13, 2026, 8:29 PM UTC

alecbz wrote: post #6

We have LLMs try to generate descriptions of PRs for us and they're pretty universally disliked. They're always overly-complex descriptions of the mechanical changes and have no sense of motivation. Also, a huge reason to understand the code yourself is to make sure the LLM isn't wrong, but this doesn't work if an LLM is itself generating the understanding.

Maybe on your team, but I don't think AI PRs are universally disliked. The people that submit PRs without understanding them are universally disliked.

Have you tried writing in AGENTS.md or whatever to exactly explain what you like/dislike about the PR descriptions?

Re: Understanding is the new bottleneck

#23

Post by stronglikedan » Thu, Aug 13, 2026, 8:29 PM UTC

bigstrat2003 wrote: post #18

Understanding has always been the bottleneck. That's why LLMs aren't actually helpful: they speed up the part which is easy (typing characters into your editor), but are neutral or even harmful on the part which is hard (understanding the problem and how best to solve it).

The part which is easy is still time consuming, so LLMs are helpful. They're just not a silver bullet.

Re: Understanding is the new bottleneck

#24

Post by elendilm » Thu, Aug 13, 2026, 8:30 PM UTC

Understanding is expensive. And hence valuable.

LLMs usually points to the most idiotic future trajectory on my work, and I have to curse it inorder to let it keep up with my refined understanding.

But what else would one expect from a probabilistic weighted next token predictor, other than to conduct probabilistic search which are 99.99% deadends.

But LLMs can pave the way towards constructing resilient and correct architecture which can be iterated fast by a human.

Architecture and determinism is where my money is in.

Re: Understanding is the new bottleneck

#25

Post by stronglikedan » Thu, Aug 13, 2026, 8:31 PM UTC

ch4s3 wrote: post #12

jbdamask wrote: post #7

pfft...I'm way past understanding

Overstanding is the new horizon

That's already been co-opted by the sovereign citizen movement, so use with caution lest you be judged for it.

Re: Understanding is the new bottleneck

#26

Post by alecbz » Thu, Aug 13, 2026, 8:35 PM UTC

dylan604 wrote: post #20

alecbz wrote: post #16

Earlier quoted context omitted.

LLM-generated (or at least LLM-assisted), but a human's still the one submitting the PR for review.

A PR with a minimal title and empty description should be refused at submission. If the human is so disinterested that they're using LLM generated code and then can't explain the purpose, that human should be prevent from making the PR. Working as a solo dev, it is very easy to be lazy like that, and I'm as guilty as anyone. Working in teams with actual reviews should absolutely have much more strict policies of what…

I mean, the “meta” right now is trying to automate as much of the software engineering flow as possible, including writing PR descriptions.

They’re shooting for LLMs being able to one-shot PRs or need minimal oversight. But yeah, in practice LLMs are not there IME.

Re: Understanding is the new bottleneck

#27

Post by dtkav » Thu, Aug 13, 2026, 8:37 PM UTC

I've been using Geoffrey's /explain-diff skill in my replace-github-with-tailor-fit-personal-software journey, and I'm liking it. I recommend at least giving it a try.

Re: Understanding is the new bottleneck

#28

Post by bckr » Thu, Aug 13, 2026, 8:39 PM UTC

avaer wrote: post #22

alecbz wrote: post #6

We have LLMs try to generate descriptions of PRs for us and they're pretty universally disliked. They're always overly-complex descriptions of the mechanical changes and have no sense of motivation. Also, a huge reason to understand the code yourself is to make sure the LLM isn't wrong, but this doesn't work if an LLM is itself generating the understanding.

Maybe on your team, but I don't think AI PRs are universally disliked. The people that submit PRs without understanding them are universally disliked. Have you tried writing in AGENTS.md or whatever to exactly explain what you like/dislike about the PR descriptions?

The basics are always so basic yet so necessary. Thanks for the recommendation.

Re: Understanding is the new bottleneck

#29

Post by jollyllama » Thu, Aug 13, 2026, 8:40 PM UTC

vjvjvjvjghv wrote: post #8

alecbz wrote: post #6

We have LLMs try to generate descriptions of PRs for us and they're pretty universally disliked. They're always overly-complex descriptions of the mechanical changes and have no sense of motivation. Also, a huge reason to understand the code yourself is to make sure the LLM isn't wrong, but this doesn't work if an LLM is itself generating the understanding.

I don’t know. From my experience I get pretty good descriptions of PRs if I ask the right questions and provide some context.

> provide some context

That's pretty much what a PR description is.

Re: Understanding is the new bottleneck

#30

Post by alecbz » Thu, Aug 13, 2026, 8:42 PM UTC

avaer wrote: post #22

alecbz wrote: post #6

We have LLMs try to generate descriptions of PRs for us and they're pretty universally disliked. They're always overly-complex descriptions of the mechanical changes and have no sense of motivation. Also, a huge reason to understand the code yourself is to make sure the LLM isn't wrong, but this doesn't work if an LLM is itself generating the understanding.

Maybe on your team, but I don't think AI PRs are universally disliked. The people that submit PRs without understanding them are universally disliked. Have you tried writing in AGENTS.md or whatever to exactly explain what you like/dislike about the PR descriptions?

The PR descriptions are pretty universally disliked. We have centralized tooling that manages the prompts for that, I’m sure they’ve tried tuning it but maybe there’s more they could do.

Though I have some local workflows where I try to teach Claude about my writing style preferences via skills and examples, and it’s still not great.