You're telling me I need to use 100% of my brain, reasoning power, and time to go over your code, but you didn't feel the need to hold yourself to the same standard?
It’s strange how AI style is so easy to spot. If LLMs just follow the style that they encountered most frequently during training, wouldn’t that mean that their style would be especially hard to spot?
It's not because they can't write PRs indistinguishable from humans, or can't write code without Emojis. It's because they don't want to freak out the general public so they have essentially poisoned the models to stave off regulation a little bit longer.
That's a lot of expensive work they're doing, and ignoring, if they're just later poisoning the models!
I'm like "Sure buddy, sure. And the nanobots are in all vaccines, right?"
Of course not! They would use it to trade and would keep it concealed while throwing the public a bone with a less advanced version. Same thing applies as AGI or even as code gen gets better.
How likely is it that you saw a plane made by humans that you misclassified due to lack of data vs how likely is it that it was actually an Alien made aircraft, coincidentally working in our atmospheric conditions, gravity and drag related physics?
Guess what, it's about likeliness and you are extrapolating the wrong assumptions. Extraordinary claims require extraordinary evidence, not the other way around.
Later the cutest of the emojis paved their way into templates used by bots and tools, and it exploded like colorful vomit confetti all over the internets.
When I see this emojiful text, my first association is not with an LLM, but with a lumberjack-bearded hipster wearing thick-framed fake glasses and tight garish clothes, rolling on a segway or an equivalent machine while sipping a soy latte.
Jk, your comments don't seem at all to me like AI. I don't see how that could even be suggested
Your comment however is just an ad hominem.
Glasses: check (I'm old)
Garish clothes: check
Segway: nope
So there's a 75% chance I am a Millenial hipster. Soy latte: sounds kinda nice
Or someone made a call that emoji-infested text is "friendlier" and tuned the model to be "friendlier."
(I once got that feedback from someone in management when writing a proposal...)
So it's entirely possible that training in one area (eg: Reddit discourse) might influence other areas (such as PRs)
But maybe it originated somewhere else.. In Javascript libraries..?
Why do you think that? I try to stay involved in accessibility community (if that's what you mean by inclusive?) and I've not heard anyone advocate for emojis over text?
I say "meme" because I believe this is how the information spreads — I think people in that particular clique suggest it to each other and it becomes a form of in-group signalling rather than an earnest attempt to improve the accessibility of information.
I'm wary now of straying into argumentum ad ignorantiam territory, but I think my observation is consistent with yours insofar as the "inclusivity" community I'm referring to doesn't have much overlap with the accessibility community; the latter being more an applied science project, and the former being more about humanities and social theory.
First, LLM style did not even exist, it's a match of several different styles, choice of words and phrases.
Second, LLM has turned a slight plurality into a 100% exclusivity.
Say, there are 20 different choices to say the same thing. They are more or less evenly distributed, one of them is a slightly more common. LLM chooses the most common one. This means that
situation before : 20 options, 5% frequency each
situation now : 1 option, 100% frequency
I think these 2 theories explain how can LLM both sound bad, and "be the most common stye, how humans have always talked" (it isn't).
Also, if the second theory is true, that is, LLM style is not very frequent among humans, that means that if you see someone on the internet that talks like an LLM, he probably is one.
God Bless.
But agree excessive emoji's, tables of things, and just being overly verbose are tells for me anymore.
The „average“ style, from the Unix manpages from the 1960s through the Linux Documentation Project all the way to the latest super-hip JavaScript isEven emoji vomit README must still have been relatively tame I assume.
They didn't learn how to write PRs. They "learned" how to write text.
Just like generic images coming out of OpenAI have the same style and yellow tint, so does text. It averages down to a basic tiktok/threads/whatever comment.
Plus whatever bias training sets and methodology introduced
Generally small startups after initial pmf. I have no idea how to run a big company and pre pmf Im guilty of "all cowboy, all the time" - YMMV
1. What is this change supposed to do?
2. Why is this change needed?
3. How was it tested?
4. Is there anything else reviewers should know?
5. Link to issue:
There's no "What changed?" because that's the diff. Explain your intent, why you think it's a good idea, how you know you accomplished your intent, and any future work needed or other concerns noticed while making the change. PR descriptions suffer from the same problem as code comments by beginners: they often just describe the "what" when that's obvious from the code, when the "why" is what's needed. So try very hard to avoid doing that.
i++; // increment i (by 1)
Doing a blame on a file, or just looking at the diff of the pull request gives you that. The why is lost very fast. After a few months it is possible that the people that did the change is not anymore in the company, so nobody to ask why something was done.
"Oh, they changed the algorithm to generate random numbers". I can see that in the code. "Why was it changed?". I have not clue if there is no extra information somewhere else like a change log, pull request description, or in the commit comments.
But all this depends on the company and size of the project. In your situation may be different.
What is unspoken here is that some open projects are using cost of submission AND cost of change / contrib as a kind of means of keeping review work down.
Nobody is correct here really. It's just that the bottlenecks have changed and we need to rethink everything.
Changing something small on a very large project is a good test. A user might simply want a new optional argument or something. Now they can do it and PR. But the process is geared towards people who know the project better even if the contributor can run all the tests it is still not trivial to fill in the PR request for a trivial change.
We need to rethink this regime shift a bit.
// loop over list and act on items
for each _, item := range items {
item.act()
}Code review is one of the places where experience is transferred. It is disheartening to leave thoughtful comments and have them met with "I duno. I just had [AI] do it."
If all you do is 'review' the output of your prompting before cutting a CR, I'd prefer you just send the prompt.
Almost nobody uses it for that today, unfortunately, and code reviews in both directions are probably where the vast majority of learning software development comes from. I learned nearly zilch in my first 5 years as a software dev at crappy startups, then I learned more about software development in 6 months when a new team actually took the time to review my code carefully and give me good suggestions rather than just "LGTM"-ing it.
Firstly, we get important benefits even when there's nothing to talk about: we get to see what the other person is working on, which stops us getting siloed or working alone. Secondly, we do leave useful feedback and often link to full articles explaining concepts, and this can be a good enough explanation for the PR author to just make the requested change. Thirdly, we escalate things to in-person discussion when appropriate, so we end up having the most valuable discussions anyway, which are around architecture, ongoing code style changes, and teaching/learning new things.
I don't understand how someone could think that async code review has almost zero value unless they worked somewhere with a culture of almost zero effort code reviews.
Sometimes a PR either merits limited input or the situation doesn't merit a thorough and thoughtful review, and in those cases a simple "lgtm" is acceptable. But I don't think that diminishes the value of thoughtful non-in-person code review.
Which is awesome and essential!
But the reason that the value of code reviews drops if they aren't done live, conducted by the person whose code is being reviewed, isn't related to the quality of the feedback. It's because a very large portion of the value of a code review is having the dev who wrote the code walk through it, explaining things, to other devs. At least half the time, that dev will encounter "aha" moments where they see something they have been blind to before, see a better way of doing things, spot discontinuities, etc. That dev has more insight into what went into the code than any other, and this is a way of leveraging that insight.
The modern form of code review, where they are done asynchronously by having reviewers just looking at the code changes themselves, is not worthless, of course. It's just not nearly as useful as the old-school method.
But otherwise, writing code with LLM‘s is more than just the prompt. You have to feed it the right context, maybe discuss things with it first so it gets it and then you iterate with it.
So if someone has done the effort and verified the result like it‘s their own code, and if it actually works like they intended, what’s wrong with sending a PR?
I mean if you then find something to improve while doing the review, it’s still very useful to say so. If someone is using LLMs to code seriously and not just to vibecode a blackbox, this feedback is still as valuable as before, because at least for me, if I knew about the better way of doing something I would have iterated further and implemented it or have it implemented.
So I don‘t see how suddenly the experience transfer is gone. Regardless if it’s an LLM assisted PR or one I coded myself, both are still capped by my skill level not the LLMs
I’ve noticed in empirical studies of informal code review that most humans tend to have a weak effect on error rates which disappears after reading so much code per hour.
Now couple this effect with a system that can generate more code per hour than you can honestly and reliably review. It’s not a good combination.
This is not just disheartening - this should be flat out refused. I'm sensitive to issues of firing people but honestly this is just someone not pulling their weight for their job.
Then the golden age of ascii encoded source, where all was easy to change.
Now we've forgotten that lesson and changed to ascii encoded binary.
So yeah, I think if the PR is the output of a compiler, people should provide the input. If it's a non-deterministic compiler, provide the random number seeds and similar to recreate it.
Makes it a lot easier to ignore, at the very least.
$$$ trillion dollar startup idea $$$
> But once that’s done, why not?
Be honest here. I don't think you do. Just like none of us have the same understanding of the code somebody else wrote. It's just a fact that you understand the code you wrote better than code you didn't.
I'm not saying you don't understand the code, that's different. But there's a deeper understanding to code you wrote, right? You might write something one way because you had an idea to try something in the future based on an idea to had while finding some bug. Or you might write it some way because some obscure part of the codebase. Or maybe because you have intuition about the customer.
But when AI writes the code, who has responsibility over it? Where can I go to ask why some choice was made? That's important context I need to write code with you as a team. That's important context a (good) engineering manager needs to ensure you're on the right direction. If you respond "well that's what the AI did" then how that any different from the intern saying "that's how I did it at the last place." It's a non-answer, and infuriating. You could also try to bullshit an answer, guessing why the AI did that (helpful since you promoted it), but you're still guessing and now being disingenuous. It's a bit more helpful, but still not very helpful. It's incredibly rude to your coworkers to just bullshit. Personally I'd rather someone say "I don't know" and truthfully I respect them more for that. (I actually really do respect people that can admit they don't know something. Especially in our field where egos are quite high. It's can be a mark of trust that's *very* valuable)
Sure, the AI can read the whole codebase, but you have hundreds or thousands of hours in that codebase. Don't sell yourself short.
Honestly I don't mind the AI acting as a reviewer to be a check before you submit a PR, but it just doesn't have the context to write good code. AI tries to write code like a junior, fixing the obvious problem that's right in front of you. But it doesn't fix the subtle problems that come with foresight. No, I want you to stumble through that code because while you write code you're also debugging and designing. Your brain works in parallel, right? I bet it does even if you don't know it. I want you stumbling through because that struggling is helping you learn more about the code and the context that isn't explicitly written. I want you to develop ideas and gain insights.
But AI writing code? That's like measuring how good a developer is by the number of lines of code they write. I'll take quality over quantity any day of the week. Quality makes the business run better and waste fewer dollars debugging the spaghetti and duct tape called "tech debt".
If the AI writes the code, you can still understand the code, but you will never know why the code is written that way. The AI itself doesn’t know, beyond the fact that that’s how it is in the training data (and that’s true even if it could generate a plausible answer for why, if you asked it).
There needs to be some responsible entity that can discuss the decisions behind the code. Those decisions have tremendous business value[0]
[0] I stress because it's not just about "good coding". Maybe in a startup it only matters that "things work". But if you're running a stable business you care if your machine might break down at any moment. You don't want the MVP. The MVP is a program that doesn't want to be alive but you've forced into existence and it is barely hanging on
If people are letting the LLM decide how the code will be written then I think they're using them wrong and yes 100% they won't understand the code as well as if they had written it by hand.
LLMs are just good pattern matchers and can spit out text faster than humans, so that's what I use them for mostly.
Anything that requires actual brainpower and thinking is still my domain. I just type a lot less than I used to.
> I know why the LLM wrote the code that way. Because I told it to and _I_ know why I want the code that way.
> If people are letting the LLM decide how the code will be written then I think they're using them wrong
I don't actually ever feel like the LLMs help me generate code faster because when writing I am also designing. It doesn't take much brain power to make my fingers move. They are a lot slower than my brain. Hell, I can talk and type at the same time, and it isn't like this is an uncommon feat. But I also can't talk and type if I'm working on the hard part of the code because I'm not just writing.
People often tell me they use LLMs to do boilerplate. I can understand this, but at the same time it begs the question "why are you writing boilerplate?" or "why are you writing so much boilerplate?" If it is boilerplate, why not generate it through scripts or libraries? Those have a lot of additional benefits. Saves you time, saves your coworkers time, and can make the code a lot cleaner because you're now explicitly saying "this is a routine". I mean... that's what functions are for, right? I find this has more value and saves more time in the long run than getting the LLMs to keep churning out boilerplate. It also makes things easier to debug because you have far fewer things to look at.
And that's a problem. By typing out the code, your brain has time to process its implications and reflect on important implementation details, something you lose out on almost entirely when letting an LLM generate it.
Obviously, your high-level intentions and architectural planning are not tied to typing. However, I find that an entire class of nasty implementation bugs (memory and lifetime management, initialization, off-by-one errors, overflows, null handling, etc.) are easiest to spot and avoid right as you type them out. As a human capable of nonlinear cognition, I can catch many of these mid-typing and fix them immediately, saving an significant amount of time compared to if I did not. It doesn't help that LLMs are highly prone to generate these exact bugs, and no amount of agentic duct tape will make debugging these issues worthwhile.
The only two ways I see LLM code generation bring any value to you is if:
* Much of what you write is straight-up boilerplate. In this case, unless you are forced by your project or language to do this, you should stop. You are actively making the world a worse place.
* You simply want to complete your task and do not care about who else has to review, debug, or extend your code, and the massive costs in capital and human life quality your shitty code will incur downstream of you. In this case, you should also stop, as you are actively making the world a worse place.
The best time to review is when writing code.
The best time to iterate on design is when writing code.
Writing code is a lot more than typing. It's the whole chimichanga
> The only two ways I see LLM code generation bring any value to you is if
That is just an opinion.
I have projects I wrote with some help from the LLMs, and I understand ALL parts of it. In fact, it is written the way it is because I wanted it to be that way.
You do not need to fully understand large codebases to use them; this is what APIs are for. If you are adventurous, you might hunt a bug in some part of a large codebase, which usually leads you from the manifestation to the source of the bug on a fairly narrow path. None of this requires "understanding all these huge codebases". Your statement implies a significant lack of experience on your part, which makes your use of LLMs for code generation a bit alarming, to be honest.
The only people expected to truly understand huge codebases are those who maintain them. And that is exactly why AI PRs are so insulting: you are asking a maintainer to vet code you did not properly vet yourself. Because no, you do not understand the generated code as well as if you wrote it yourself. By PRing code you have a subpar understanding of, you come across as entitled and disrespectful, even with the best of intentions.
> That is just an opinion.
As opposed to yours? If you don't want to engage meaningfully with a comment, then there is no need to reply.
> I have projects I wrote with some help from the LLMs, and I understand ALL parts of it. In fact, it is written the way it is because I wanted it to be that way.
See, I could hit you with "That is just an opinion" here, especially as your statement is entirely anecdotal But I won't, because that would be lame and cowardly.
When you say "because I wanted it to be that way", what exactly does that mean? You told an extremely complex, probabilistic, and uninterpretable automaton what you want to write, and it wrote it not approximately, but exactly as you wanted it? I don't think this is possible from a mathematical point of view.
You further insist that you "understand ALL parts" of the output. This actually is possible, but seems way too time-inefficient to be plausible. It is very hard to exhaustively analyze all possible failure modes of code, whether you wrote it yourself or not. There is a reason why certifying safety-critical embedded code is hell, and why investigating isolated autopilot malfunctions in aircraft takes experts years. That is before we consider that those systems are carefully designed to be highly predictable, unlike an LLM.
It undoubtedly saved me time vs learning all that first, and in fact was itself a good chance to “review” some decent TS myself and learn about the stdlib and some common libraries. I don’t think that effort missed many critical idioms and I would say I have decent enough taste as an engineer that I can tell when something is janky and there must be a better way.
> but I’m not a TS expert
> So the most recent thing that I did
I was working on a setup.py file and I knew I had done something small and dumb, but was being blind to it. So I pulled up claude code and had it run parallel to my hunt. Asked it to run the build command and search for the error. It got caught up in some cmake flags I was passing, erroneously calling them errors. I get a number of prompts in and they're all wrong. I fixed the code btw, it was a variable naming error (classic!).
I've also had success with claude, but it is super hit or miss. I've never gotten it to work well for anything remotely complicated if there also isn't the code in a popular repo I could just copy paste. But it is pretty hit or miss for even scripts, which I write a lot of bash. People keep telling me it is great for bash and honestly guys, just read the man pages... (and use some god damn functions!)
That's why it isn't necessary to add the "to be fair" comment i see crop up every time someone complains about the low quality of AI.
Dealing with low effort people is bad enough without encouraging more people to be the same. We don't need tools to make life worse.
It's as if someone created a device that made cancer airborne and contagious and you come in to say "to be fair, cancer existed before this device, the device just made it way worse". Yes? And? Do you have a solution to solving the cancer? Then pointing it out really isn't doing anything. Focus on getting people to stop using the contagious aerosol first.
If a company builds an industrial poop delivery system that lets anyone with dog poop deliver it directly into my yard with the push of a button I have a much different and much bigger problem
Waiting for the rest of the comment to load in order to figure out if it's sincere or parody.
This is like reviewing your own PRs, it completely defeats the purpose.
And no, using different models doesn’t fix the issue. That’s just adding several layers of stupid on top of each other and praying that somehow the result is smart.
That is literally how civilization works.
As an example, knowing that a service is offered by a registered company with presence in my area gives me the knowledge "that they know that I know" that if something goes wrong, I can sue them for negligence, possibly up to piercing the corporate veil the company and having the directors serve prison time. From that I can somewhat rationally derive that if the company has been in business offer similar services for years, it is likely that they have processes in place to maintain a level of professionalism that would lower the risk of such lawsuits. And on an organisational level, even if I still have good reason to think that most of the employees are incompetent, the fact that the company is making it work gives me a significantly higher preference in the "result" than I would in any individual "stupid" component.
And for a closer-to-home example, the internet is well known to be a highly reliable system built from unreliable components.
It's a joke.
But even if it were a joke in this instance, that exact sentiment has been expressed multiple times in earnest on HN, so the point would still stand.
As insulting as it is to submit an AI-generated PR without any effort at review while expecting a human to look it over, it is nearly as insulting to not just open the view the reviewer will have and take a look. I do this all the time and very often discover little things that I didn't see while tunneled into the code itself.
In the sense that you double check your work, sure. But you wouldn’t be commenting and asking for changes, you wouldn’t be using the reviewing feature of GitHub or whatever code forger you use, you’d simply make the fixes and push again without any review/discussion necessary. That’s what I mean.
> open the view the reviewer will have and take a look. I do this all the time
So do I, we’re in perfect agreement there.
It is, but for all the reasons AI is supposed to fix. If I look at code I myself wrote I might come to a different conclusion about how things should be done because humans are fallible and often have different things on their mind. If it's in any way worth using an AI should be producing one single correct answer each time, rendering self PR review useless.
Yes. You just have to be in a different mindset. I look for cases that I haven't handled (and corner cases in general). I can try to summarize what the code does and see if it actually meets the goal, if there's any downsides. If the solution in the end turns out too complicated to describe, it may be time to step back and think again. If the code can run in many different configurations (or platforms), review time is when I start to see if I accidentally break anything.
> This is like reviewing your own PRs, it completely defeats the purpose.
I've been the first reviewer for all PRs I've raised, before notifying any other reviewers, for so many years that I couldn't even tell you when I started doing it. Going through the change set in the Github/Gitlab/Bitbucket interface, for me, seems to activate an different part of my brain than I was using when locked in vim. I'm quick to spot typos, bugs, flawed assumptions, edge cases, missing tests, to add comments to pre-empt questions ... you name it. The "reading code" and "writing code" parts of my brain often feel disconnected!
Obviously I don't approve my own PRs. But I always, always review them. Hell, I've also long recommended the practice to those around me too for the same reasons.
You don’t, we’re on the same page. This is just a case of using different meanings of “review”. I expanded on another sibling comment:
https://news.ycombinator.com/item?id=45723593
> Obviously I don't approve my own PRs.
Exactly. That’s the type of review I meant.
So, your minimum bar for a useful AI is that it must always be perfect and a far better programmer than any human that has ever lived?
Coding agents are basically interns. They make stupid mistakes, but even if they're doing things 95% correctly, then they're still adding a ton of value to the dev process.
Human reviewers can use AI tools to quickly sniff out common mistakes and recommend corrections. This is fine. Good even.
You are transparently engaging in bad faith by purposefully straw manning the argument. No one is arguing for “far better programmer than any human that has ever lived”. That is an exaggeration used to force the other person to reframe their argument within its already obvious context and make it look like they are admitting they were wrong. It’s a dirty argument, and against the HN guidelines (for good reason).
> Coding agents are basically interns.
No, they are not. Interns have the capacity to learn and grow and not make the same mistakes over and over.
> but even if they're doing things 95% correctly
They’re not. 95% is a gross exaggeration.
> This makes no sense, and it’s absurd anyone thinks it does. If the AI PR were any good, it wouldn’t need review. And if it does need review, why would the AI be trustworthy if it did a poor job the first time?
This is an entirely unfair expectation. Even the best human SWEs create PRs with significant issues - it's absurd by the parent to say that if a PR is "any good, it wouldn’t need review"; it's just an unreasonable bar, and I think that @latexr was entirely justified in pushing back against that expectation.
As for the "95% correctly", this appears to be a strawman argument on your end, as they said "even if ...", rather than claiming that this is the situation at the moment. But having said that, I would actually like to ask both of you - what does it even mean for a PR to be 95% correct - does it mean that that 95% of the LoC are bug-free, or do you have something else in mind?
ultimately I'm happy to fight fire with fire. there was a time I used to debate homophobes on social media - I ended up writing a very comprehensive list of rebuttals so I could just copy and paste in response to their cookie cutter gotchas.
The point of most jobs is not to get anything productive done. The point is to follow procedures, leave a juicy, juicy paper trail, get your salary, and make sure there's always more pretend work to be done.
That's certainly not my experience. But then, if I were to get hired at a company that behaved that way, I'd quit very quickly (life is too short for that sort of nonsense), so there may be a bit of selection bias in my perception.
I first read that as "coworkers (who are) fully AI generated" and I didn't bat an eye.
All the AI hype has made me immune to AI related surprises. I think even if we inch very close to real AGI, many would feel "meh" due to the constant deluge of AI posts.
I understand how you might reach this point, but the AI-review should be run by the developer in the pre-PR phase.
Do you review your comments too with AI?
This reminds me of an awesome bit by Žižek where he describes an ultra-modern approach to dating. She brings the vibrator, he brings the synthetic sleeve, and after all the buzzing begins and the simulacra are getting on well, the humans sigh in relief. Now that this is out of the way they can just have a tea and a chat.
It's clearly ridiculous, yet at the point where papers or PRs are written by robots, reviewed by robots, for eventual usage/consumption/summary by yet more robots, it becomes very relevant. At some point one must ask, what is it all for, and should we maybe just skip some of these steps or revisit some assumptions about what we're trying to accomplish
I've been thinking this for a while, despairing, and amazed that not everyone is worried/surprised about this like me.
Who are we building all this stuff for, exactly?
Some technophiles are arguing this will free us to... do what exactly? Art, work, leisure, sex, analysis, argument, etc will be done for us. So we can do what exactly? Go extinct?
"With AI I can finally write the book I always wanted, but lacked the time and talent to write!". Ok, and who will read it? Everybody will be busy AI-writing other books in their favorite fantasy world, tailored specifically to them, and it's not like a human wrote it anyway so nobody's feelings should be hurt if nobody reads your stuff.
In the dating scenario what's really absurd and disgusting isn't actually the artificiality of toys.. it's the ritualistic aspect of the unnecessary preamble, because you could skip straight to tea and talk if that is the point. We write messages from bullet points, ask AI to pad them out uselessly with "professional" sounding fluff, and then on the other side someone is summarizing them back to bullet points? That's insane even if it was lossless, just normalize and promote simple communications. Similarly if an AI review was any value-add for AI PR's, it can be bolted on to the code-gen phase. If editors/reviewers have value in book publishing, they should read the books and opine and do the gate-keeping we supposedly need them for instead of telling authors to bring their own audience, etc etc. I think maybe the focus on rituals, optics, and posturing is a big part of what really makes individual people or whole professions obsolete
It's 90% AI, but that 90% was almost entirely boilerplate and would have taken me a good chunk of time to do for little gain other than the fact I did it.
I don’t think they are (telling you that). The person who sends you an AI slop PR would be just as happy (probably even happier) if you turned off your brain and just merged it without any critical thinking.
I can see the code, I know what changed. Give me the logic behind this change. Tell me what issues you ran into during the implementation and how you solved them. Tell me what other approaches you considered and ruled out.
Just saying "This change un-links frobulation from reticulating splines by doing the following" isn't useful. It's like adding code comments that tell you what the next line does; if I want to know that I'll just read the next line.
The AI has far more energy than I do when it comes to writing PR summaries, I have done it so many times, it's not the main part of my job. I have already provided all the information for a PR, why should I repeat myself? What happened to DRY?
A good PR summary should be the why of the PR. Not redundantly repeat what changed, give me description of why it changed, what alternatives were tested, what you think the struggles were, what you think the consequences may be, what you expect the next steps to be, etc.
I've never seen an AI generated summary that comes close to answering any of those questions. An AI generated summary is a bit like that junior developer that adds plenty of comments but all the comments are:
// add x and y
var result = x + y;
I'm going to read the code anyway to review a PR, a summary of what the code already says it does is redundant information to me.
No, they're expecting you to half-ass it too, and just use AI to review the PR.
I have worked at smaller firms, mostly, early stage (< 50 engineers), and folks are super busy. Having AI support in writing better thoughtful commentary, provide deeper context is a boon.
In the end, I'll have to say "it depends" -- you can't just throw slop at people but there's definitely a middle ground where everyone wins.
# Minimal Reprex (Correct)
(unintelligible nonsense here)
And here is the correct, minimal fix, guaranteed to work:
# Correct Fix (Correct)
(same unintelligible nonsense, wrapped in a try/catch block)
Make this change and your code should work perfectly!