Hacker Newsnew | past | comments | ask | show | jobs | submit | pinkgolem's commentslogin

What really important things are human reviews catching in your org?

I just feel more and more like the effort invested in manual reviews is not worth it


1. Whether the thing should be done in the first place

2. If it's the correct solution on a high level

3. Whether it conflicts with or duplicates other parts of the system

4. Whether the comments are actually useful or restating the LLM chat

Also many others but these are the most common IME


All of these are angles an AI reviewer can test for as well, and will (IME) mostly catch mistakes correctly. I also still manually review code, and usually also catch issues, but the severity of what I find shrinks ever further as agents get better.

The sprawling code comments are becoming the most draining part of code review though, that's really killing me from the inside.


> All of these are angles an AI reviewer can test for as well, and will (IME) mostly catch mistakes correctly.

No, none of today's AI would give you enough signal around "should this thing be built in the first place" nor if it's the correct solution on a high level.

They don't understand why you are doing what you are doing, and even if you explain it, they still don't actually understand the motivation and lots of other things.

You'll get them to do guesses and pretend they actually know how to prioritize and will tell you it makes lots of sense, whatever they come up with. But try following it blindly and you'll see where you end up.

This is why "one agent + one good developer" beats "thousands of agents working in a swarm" still today.


I don’t think I claimed agent reviews to be a panacea. It’s a tool that can help you lower the review pressure in companies working with agentic coding tools.

Someone asked:

> What really important things are human reviews catching in your org?

Another person said:

> 1. Whether the thing should be done in the first place

And you replied:

> All of these are angles an AI reviewer can test for as well, and will (IME) mostly catch mistakes correctly

Which as I noted, is very far from the truth. I neither claimed that you said "agent reviews are a panacea", but when you claim "AI can solve all those things" and two of the first items cannot be addressed by AI (today), then I'm rebuking those specific things, not some other general point you implicitly made.


But that’s the thing, when I say "mostly" that sure doesn’t imply it can solve all those things - it can help to a great extent.

It is mostly useless at figuring out if "should this thing be built in the first place" and "if it's the correct solution", and mostly cannot help at all with those things.

Where "mostly" means kind of what it says but also not really.


The AI review are still quite far from having the same level of critical thinking and high level knowledge of your application, what you have done in the past and want to do next etc.

If you don't master this for your own project, what's even the point of your job.


I would assume 1/2 are discussed beforehand?

1. Does the implementation fit in the architecture/style of the project?

2. Are there potential security, accessibility, performance, etc. issues?

3. Domain specific knowledge (SQL, ASP.NET, XQuery, etc.) where there are better ways of solving a problem, or possible issues not handled.

4. Sense checking ... is the code easy to read? does it need an explanatory comment? does it need named parameters? etc.


The free one in Google has major hallucinations..


deepseek is a lab, the models are open weight(atleast after a bit)

they have a somewhat selfhostable model(flash), but are mostly known for having super cheap api access

no zdr ofcourse.


> no zdr ofcourse.

Since the models are open, there's other infra providers who have different policies.


Yes, but those are more expensive normally


any recs?


Tbh I just go through openrouter because it's easy, so you can just go down the list in that and see who sounds good. I honestly don't spend much time at it.


i see. appreciate the reply this was concise and just what i needed


Open ai is allowing subscription use, anthropic also paused the effort to stop subscription use.


They did? Is there a source where I can learn more? I'd love to use my Anthropic subscription with opencode.


Only if opencode uses the agent sdk/Claude -p

June 15 changes would be the keyword to check, but at least in anthropics case it's... Complicated


Bans are not in effect?


You can also change the writing style with a prompt quit easily, if the same person would publish millions of articles, we would also recognize them.


The Turing Test is about being able to figure out if “someone” is a computer during a short conversation, not “millions of articles”.

LLMs still live in the uncanny valley and can be sussed out immediately.

For example, I’m extremely annoyed by the fact that offshore developers respond to me almost exclusively using text generated by Claude. You can tell immediately because they use overly descriptive techno word salad that no normal human uses unless they are trying to be ultra specific for a scientific paper - and even then it’s still too much for a real person.


My argument is, you recognize the style because you know it..

5 years ago, you would not have been able to determine that this was the case, and just have assumed it's a know it all character.

Tell your offshore developers to use caveman or so


I don't understand. do you think that people can't figure it out if they speak with llms or not from a few turns?


We don't train current LLMs to mimic the average human's writing style. We train it to be smart, helpful, and knowledgable, too knowledgeable for a human. We can easily train a LLM to pass the Turing test if we wanted to, but then it would just sound dumb or biased.


> We can easily train a LLM to pass the Turing test if we wanted to, but then it would just sound dumb or biased.

Interesting idea. "This is not dumb and biased enough, probably not a human".


There is tons of money trying to get customer support bots to sound human but they fool nobody.


> but it would just sound dumb or biased

We have one of those: Grok.


Yes, and it's indistinguishable from the average X user.


Bold of you to assume they’re actual users.


Does it? I found llms to be great for straight conversions

At least if tescoverage is good, but well... That's something llms can also be used for


You can only outlaw stuff in your own countrys..

And if cheaper access is an advantage, other countries will surpass you


I mean... I need ~120 watt an hour to keep the house between 20-23 C.


In Nevada?


Fair, Germany

Only over 35C

But it's not even a well isolated house.


I mean, looking at most us company's.. What support?


Have you tried cnpg? Worked amazingly well for my usecases


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: