ptnpzwqd

4 ms·

1.
by ptnpzwqd 3mo ago
Be that as it may, it would seem absurd if we start calling distillation out as antagonistic, but don't do the same for the SOTA models being trained on human-created data.
2.
by ptnpzwqd 4mo ago
This does not really match my observations. While it does feel to me the sentiment is shifting towards a more negative one, overall HN feels reasonably balanced between those that are pro-AI and those anti-AI (with the middle ground somewha
3.
by ptnpzwqd 5mo ago
Responses seem to be very "either or" as usual on such topics. I think it should be possible to appreciate how impressive this is on one hand, while also discussing the limitations of the approach. Everyone can probably agree that
4.
by ptnpzwqd 5mo ago
I think a notable difference is that the AI that is portrayed in most sci-fi (that I have read/watched anyway) tend to be "logical machines" that act deterministically based on the data available to them. What we got are &quo
5.
by ptnpzwqd 6mo ago
I feel the conclusions here are a bit thin. Code quality tends to have an impact on more than just aesthetics - and Claude Code certainly feels like a buggy mess from an end user's perspective. Of course people still use Claude Code, b
6.
by ptnpzwqd 7mo ago
On the falling behind: I strongly doubt that is going to be the case - picking up these tools is not rocket science, even if you want to be able to use them fairly effectively. In addition, there is so much churn in AI tooling these days th
7.
by ptnpzwqd 7mo ago
I live in two realities too. One where articles like this talk about a 10x increase in productivity, where the sentiment on HN is that all software can now be vibe coded without review, where AGI is right around the corner. And another real
8.
by ptnpzwqd 7mo ago
It was maybe not quite clear enough in my comment, but this is more of a hypothetical future scenario - not at all where I assess LLMs are today or will get to in the foreseable future. So it becomes a bit theoretical, but I guess if we had
9.
by ptnpzwqd 7mo ago
Yes, but LLM-based reviews are not nearly a compensation for human review, so it doesn't change much.
10.
by ptnpzwqd 7mo ago
Yes, I agree. It was just me playing with a hypothetical (but in my view not imminent) future where vibe-coding without review would somehow be good enough.
11.
by ptnpzwqd 7mo ago
At the moment verification at scale is an unsolved problem, though. As mentioned, I think this will act as a rough filter for now, but probably not work forever - and denying contributions from non-vetted contributors will likely end up bei
12.
by ptnpzwqd 7mo ago
Sure - and I suspect we will see that soon enough. But it has downsides too, and finding the right way to vet potential contributors is tricky.
13.
by ptnpzwqd 7mo ago
Even if we assume that LLMs become good enough for this to be true (some might feel that is the case already - I disagree, but that is beside the point), there is no reason why OSS maintainers should accept such outside contributions that t
14.
by ptnpzwqd 7mo ago
The problem is the increasing review burden - with LLMs it is possible to create superficially valid looking (but potentially incorrect) code without much effort, which will still take a lot of effort to review. So outright rejecting code t
15.
by ptnpzwqd 7mo ago
Not necessarily a bad idea, but I think the bigger issue here and now is the increasing assymmetry in effort between code submitter and reviewer, and the unsustainable review burden on the maintainers if nothing is done.
16.
by ptnpzwqd 7mo ago
I suspect this is for now just a rough filter to remove the lowest effort PRs. It likely will not be enough for long, though, so I suspect we will see default deny policies soon enough, and various different approaches to screening potentia
17.
by ptnpzwqd 7mo ago
I think this is a reasonable decision (although maybe increasingly insufficient). It doesn't really matter what your stance on AI is, the problem is the increased review burden on OSS maintainers. In the past, the code itself was a sor
18.
by ptnpzwqd 7mo ago
Correct, but that has and probably always will be the case. You spend the time on what is needed for you to move ahead - if code review is now the most time consuming part, that is where you will spend your time. If ever that is no longer a
19.
by ptnpzwqd 7mo ago
I of course cannot say what the future holds, but current frontier models are - in my experience - nowhere near good enough for such autonomy. Even with other agents reviewing the code, good test coverage, etc., both smaller - and every now
20.
by ptnpzwqd 7mo ago
If reviewing has become the bottleneck, the obvious - albeit slightly boring - solution is to slow down spitting out new code, and spend relatively more time reviewing. Just going ahead and piling up PRs or skipping the review process is of
21.
by ptnpzwqd 7mo ago
You can use this: hnthrowaway.outboard407@passmail.net
22.
by ptnpzwqd 7mo ago
I don't expect that - am merely responding to the parent comments claim that Claude consistently one-shots production ready code (which does not at all match my observations).
23.
by ptnpzwqd 7mo ago
Feel free to share it, would be very curious - ideally alongside the prompts.
24.
by ptnpzwqd 7mo ago
I have used Claude (incl. Opus 4.6) fairly extensively, and Claude still spits out quality that is far below what I would call production ready - both littered with smaller issues, but also the occasional larger blunder. Particularly when d
25.
by ptnpzwqd 7mo ago
There are some really baffling takes here. And it doesn't really matter how good or bad coding agents are. Coding agents greatly reduce the barrier to contributing something that at least looks okay at the surface, so reviewing contrib
26.
by ptnpzwqd 7mo ago
This article seems completely out of line with reality, maybe I am living on a different planet. I have never heard of anyone following those SDLC steps rigorously and sequentially. Things tend to be much more intertwined, combined, and ite
27.
by ptnpzwqd 3y ago
This makes the assumption that the increase in headcount was to cover the surge in demand, and the additional assumption that the additional headcount was needed to satisfy the demand (obviously, some of it might). For technology in particu