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

The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

avikalpg.github.io

Post reply on HN
Print view View original

21–30 of 35 posts

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#21

Post by donmcronald »

CompoundEyes wrote: post #17
donmcronald wrote: post #13

Earlier quoted context omitted.

How do you make it deterministic?

Sorry I meant that it's javascript / typescript so we can deterministically orchestrate a series of prompts and shape their output exactly how we'd like. Returning the review as structured output as a JSON object is very helpful for this. If the review result seems bungled, run a judge prompt at the end and tell it to go try again ^_^.

That makes sense. You could even run against multiple models or future models then, right? I can see some value in that because maybe 2 years from now the models will be able to surface issues that weren’t detected originally. I suppose you could run against the whole codebase in the future, but could also imagine something that could track down where a bug was introduced.

Do you save the reviews or discard them?

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#22

Post by fourside »

Boldened15 wrote: post #18
Legend2440 wrote: post #5

Man, I hate when I'm reading a blogpost and then I realize the whole thing is just an ad for a startup.

I'm curious, why? The startup connection is disclosed in the third sentence so it's not particularly hidden. And at least if it's a real product they're putting their money where their mouth, you can check out their product or reviews from customers to see if their approach is right. Someone working on AI tooling for code reviews is exactly the right person I'd want to get an opinion from on the space, otherwise it's…

Because someone selling AI tooling is incentivized to sell you AI, not to inform you.

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#23

Post by HoyaSaxa »

CompoundEyes wrote: post #12

I put in a code reviewer that runs and comments when a pull request is created using Github actions and Microsoft GenAIScript. It's pretty straightforward. The key thing is we have total control over the prompt to fit our repo and devs needs, can make it multi-stage and deterministic using Typescript code or use agents in GenAIScript to open adjacent files for more context. The value we've received is that a dev can…

I’d be interested in seeing the scripts if you are able to share (redacted) versions of them

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#24

Post by Herring »

mrshadowgoose wrote: post #20
Herring wrote: post #19

Earlier quoted context omitted.

Billions of dollars from Google, basically. https://www.businessinsider.com/robotaxis-may-mobility-tesla... High operational costs, low revenue potential, technical difficulties, competitors exiting the space.

Sorry, that's goalpost moving. Just reminding you of your earlier claim: > AI is not good enough yet for anything requiring deep reasoning, mission-critical work... Is driving a mission-critical function? Due to its safety critical nature, many would say "yes". So have you simply pivoted to "oh it does work, but it's not as profitable as it should be"?

You started the business discussion, not me. You need to decide if you want to continue it or not.

When your tech is unreliable, it costs you money. The need for remote human assistance during edge cases means you have to staff that. You need really expensive sensors, so more upfront costs and maintenance. You can't run when it's rainy/snowy/etc, so more downtime. Maybe even slower rides and longer wait times so lower utilization. And of course more AI updates, more training, more data cleaning/labeling, more engineers etc etc.

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#25

Post by layer8 »

lukaslalinsky wrote: post #2

Is the purpose of these tools really to spend less time? I think their main value is reducing mistakes through having one extra set of eyes, even if mechanical ones, looking at the code. As a sole developer of a non-trivial open source project, I've recently started using CodeRabbit, very skeptical about it, but right on the first PR, it actually found a bug that my CI tests did not catch, decided to keep it after th…

These code review tools are basically analogous in function to a linter. They flag potential issues, but you still have to check all of them.

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#26

Post by mrshadowgoose »

Herring wrote: post #24
mrshadowgoose wrote: post #20

Earlier quoted context omitted.

Sorry, that's goalpost moving. Just reminding you of your earlier claim: > AI is not good enough yet for anything requiring deep reasoning, mission-critical work... Is driving a mission-critical function? Due to its safety critical nature, many would say "yes". So have you simply pivoted to "oh it does work, but it's not as profitable as it should be"?

You started the business discussion, not me. You need to decide if you want to continue it or not. When your tech is unreliable, it costs you money. The need for remote human assistance during edge cases means you have to staff that. You need really expensive sensors, so more upfront costs and maintenance. You can't run when it's rainy/snowy/etc, so more downtime. Maybe even slower rides and longer wait times so lowe…

> You can't run when it's rainy/snowy/etc, so more downtime.

Waymo vehicles don't work in the rain? This is easily verifiable as false.

You really have a tough time with being wrong, don't you?

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#27

Post by avikalp »

Legend2440 wrote: post #5

Man, I hate when I'm reading a blogpost and then I realize the whole thing is just an ad for a startup.

I'm sorry, I didn't mean it to be an ad. I have been interviewing engineering leaders for months, and my startup idea is born out of it. I don't have the product ready yet - it is evolving based on what I am learning.

I just thought it would be a good idea to share what I have learnt.

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#28

Post by avikalp »

Herring wrote: post #6

AI is not good enough yet for anything requiring deep reasoning, mission-critical work, error detection at a human-expert level, or handling unpredictable edge cases. It just talks like it's very smart, and humans apparently have a bias for persuasive communication skills. It's also very fast, which humans also think indicates general intelligence. But it's not, and that's why most LLM tools are author-focused, so th…

I have had a similar discussion with a fellow On-Deck Founder, and here is where we reached:

- More than being "good enough", it is about taking responsibility. - A human can make more mistakes than an AI, and they are still the more appropriate choice because humans can be held responsible for their actions. AI, by its very nature, cannot be 'held responsible' -- this has been agreed upon based on years of research in the field of "Responsible AI". - To completely automate anything using AI, you need a way to trivially verify whether it did the right thing or not. If the output cannot be verified trivially, you are just changing the nature of the job, and it is still a job or a human being (like the staff you mentioned who remotely control Waymos when something goes wrong). - If an action is not trivially verifiable and requires AI's output to directly reach the end-user without a human-in-the-loop, then the creator is taking a massive risk. Which usually doesn't make sense for a business when it comes to mission-critical activities.

In Waymo's case, they are taking massive risks because of Google's backing. But it is not about being 'good enough'. It is about the results of the AI being trivially verifiable - which, in the case of driving, is true. You just need three yes/no answers: Did the customer reach where they wanted? Are they safe? Did they arrive on time? Are they happy with the experience?

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#29

Post by avikalp »

lukaslalinsky wrote: post #2

Is the purpose of these tools really to spend less time? I think their main value is reducing mistakes through having one extra set of eyes, even if mechanical ones, looking at the code. As a sole developer of a non-trivial open source project, I've recently started using CodeRabbit, very skeptical about it, but right on the first PR, it actually found a bug that my CI tests did not catch, decided to keep it after th…

What you are saying is true, and this is the feedback I hear every time I talk to a small team of developers (generally fewer than 15 developers).

At this stage, you don't need "another set of eyes" because it is not that big of a problem to break something, as you are not going to lose massive amounts of money because of the mistake.

All these teams need is a sanity check. They also generally (even without the AI code reviewers) do not have a strong code review process.

This is why, in the article, I have clearly mentioned that these are learning based on talking to engineers in Series-B and Series-C startups.

Re: The AI Code Review Disconnect: Why Your Tools Aren't Solving Your Real Problem

#30

Post by avikalp »

savanaly wrote: post #3

The thing is that inserting AI into the code reviewer side doesn't make too much sense. Unless they have a different AI doing the reviewing than the one that helped to write it, there won't be anything left to say at that stage. The AI was already involved in writing it and as they mention in the article there's points in the writing-with-AI process where the AI editor will try to catch bugs, educate the developer, a…

I agree with you 100%.

In the maker-checker process, if we are imagining a future where AI will be writing/editing most of the code, the AI-code-review tools will need to integrate within its agentic process.

And the job of a better code-review interface (like the one that I am trying to build) would be to provide a higher level of abstraction to the user so that they can verify the output of the AI code generators more effectively.

Post reply on HN