[HN Gopher] Why Greptile just does code reviews and doesn't also...
___________________________________________________________________
Why Greptile just does code reviews and doesn't also generate code
Author : dakshgupta
Score : 34 points
Date : 2025-08-04 14:43 UTC (8 hours ago)
(HTM) web link (www.greptile.com)
(TXT) w3m dump (www.greptile.com)
| o11c wrote:
| This is advertising for an AI product. Slightly more interesting
| background story than most articles doing so, but still an ad for
| a product that probably won't actually work.
| mooreds wrote:
| I have no direct experience with Greptile, but asked about AI
| code review assistants on a mailing list of engineering leaders
| I'm on. Several folks suggested greptile.
|
| So, consider this hearsay that it works.
| vouwfietsman wrote:
| I was so hoping that this would not be about AI, and actually
| talk about how we need to do better as an industry and start
| using objective measures of software quality backed by government
| standards.
|
| Nope, its about AI code reviewing AI, and how that's _a good
| thing_.
|
| Its like everyone suddenly forgot the old adage: "code is a
| liability".
|
| "We write code twice as fast!" just means "we create liability
| twice as fast!". It's not a good thing, at all.
| dang wrote:
| I had thought that putting quotes around the phrase
| "independent auditor" above would have prevented this sort of
| misunderstanding, but clearly not, so I've changed the title to
| something more straightforward now.
|
| (Submitted title was "Software needs an independent auditor")
| literalAardvark wrote:
| Improving automated code review does improve software, so idk
| why you're grinding that axe in this particular thread.
| brynary wrote:
| This rings similar to a recent post that was on the front page
| about red team vs. blue team.
|
| Before running LLM-generated code through yet more LLMs, you can
| run it through traditional static analysis (linters, SAST, auto-
| formatters). They aren't flashy but they produce the same results
| 100% of the time.
|
| Consistency is critical if you want to pass/fail a build on the
| results. Nobody wants a flaky code reviewer robot, just like
| flaky tests are the worst.
|
| I imagine code review will evolve into a three tier pyramid:
|
| 1. Static analysis (instant, consistent) -- e.g using Qlty CLI
| (https://github.com/qltysh/qlty) as a Claude Code or Git hook
|
| 2. LLMs -- Has the advantage of being able to catch semantic
| issues
|
| 3. Human
|
| We make sure commits pass each level in succession before moving
| on to the next.
| dakshgupta wrote:
| Reading that post sent me down the path to this one. This stack
| order makes total sense, although in practice it's possible 1-2
| merge into a single product with two distinct steps.
|
| The 3. is interesting too - my suspicion is that ~70% of PRs
| are too minor to need human review as the models get better,
| but the top 30% will because there will be opinion on what is
| and isn't the right way to do that complex change.
___________________________________________________________________
(page generated 2025-08-04 23:01 UTC)