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

There is an AI code review bubble

greptile.com

Post reply on HN
Print view View original

261–265 of 265 posts

Re: There is an AI code review bubble

#261

Post by jv22222 »

just6979 wrote: post #247
jv22222 wrote: post #226

We built an internal code review tool at the day job and are getting pretty good results with it (CLI tool). Here's a summary of the top-level ideas behind it. Hope it's helpful! Core Philosophy - "Advisor, not gatekeeper" - Every issue includes a "Could be wrong if..." caveat because context matters and AI can't see everything. Developers make the final call. (Just this idea makes it less annoying and stops devs goi…

If you need the AI to indicate "could be wrong" on everything it writes to prevent your devs from blindly following everything it says, you're doing it so wrong. That should be the default mindset. Of course it could be wrong.

Not quite.

The could be wrong part is very helpful because it has (on multiple occasions) dug up something that was long-lost-to-lore about why something should work in a non conventional way.

Without that, the advice looks perfectly sensible and would send devs down a Rabbit hole, because the AI recommendation "looks right".

Re: There is an AI code review bubble

#264

Post by DavidYoussef »

The article nails the core issue but I think misdiagnoses the solution space.

The problem isn't that AI code review exists - it's that current tools are solving the wrong problem. They review code that humans wrote. The actual crisis is reviewing code that AI wrote.

  When AI increases code volume by 10x but reviewer count stays flat, you
  don't need better review tools. You need risk triage. Not every PR deserves
  the same attention:

  - Typo fix to a README? L0. Auto-approve with an evidence log.
  - New utility function with tests? L1. One model scans it, posts findings.
  - Changes to auth middleware or payment flow? L3. Three models have to reach
    consensus before a human even looks at it.
  - Production deployment config? L4. Models + mandatory human sign-off.

  We've been building this (codeguard-action on GitHub, MIT licensed) - a
  GitHub Action that classifies PR risk, runs multi-model review proportional
  to that risk, and produces a cryptographic evidence bundle proving what was
  checked. The evidence is hash-chained and independently verifiable offline
  with a separate tool.

  The point isn't to replace human reviewers. It's to stop burning them out
  on L0-L1 changes so they have capacity for the L3-L4 ones that actually
  matter.

  The 786-PR-backlog problem mentioned upthread isn't a review problem. It's
  a triage problem.

Re: There is an AI code review bubble

#265

Post by atomicnature »

AI code review has genuinely helpful - especially when we generate code with copilot, etc.

Many times, these GenAI tools can delete/modify code mistakenly.

I use LiveReview's git precommit features - so the review happens right before I commit code automatically. And it has saved me many (100s of) times.

Give LiveReview's Precommit checks a try.

Post reply on HN