Code Review Challenges - Harness engineering: Leveraging Codex in an agent-first world

Ralph Wiggum Loop concept, agents reviewing their own code, human inability to meaningfully review hundreds of AI-generated lines, comparisons to factory workers testing products

Comparing the review of AI-generated code to factory workers frantically puffing on handfuls of e-vapes to test them, commenters suggest that the overwhelming volume of output often forces humans into superficial or selective checks. To mitigate this, some developers employ a "Ralph Wiggum Loop" where multiple agents recursively review and refine their own code, though critics warn this can become an expensive, inefficient cycle resembling a "snake eating its own tail." Ultimately, while new tools and "skill" files attempt to automate quality control, many practitioners argue that the persistent tendency of models to produce "bad hacks" still necessitates rigorous, line-by-line human steering.

View on HN · Topics

The other day I came across to a video showing workers in a e-vape factory. They pick up a bunch of e-vapes from the conveyor belt (each has 6 e-vape think), stick in their mouth and vigorously vape all of them for about 5 seconds, then test the next bunch. Humans reviewing hundreds of lines of change in a PR written by AI is not very different.

View on HN · Topics

Very true. If a PR has 1000 lines I would check only a handful full of them and leave the rest for test suit .

View on HN · Topics

I agree with many of the points made by nimonian above (esp the one starting with 'make a single skill called "code" which describes the lifecycle'), based on my limited experience with these things.

I'm building a skill + CLI tool along those lines (for solo devs not corporates). Here is what my "lifecycle" type skill looks like right now: https://github.com/bitkentech/shipsmooth/blob/releases/dist/... (warning, heavily work in progress). You can see a demo here: https://shipsmooth.net/

I was not happy with the default code quality generated by Claude Code. So I've been adding some skill-file rules to address that, and so far happy with the results: https://github.com/bitkentech/shipsmooth/tree/main/skills/ex... . There was a similar one on HN yesterday called opencodereview: https://news.ycombinator.com/item?id=48406358

There are many such workflows out there! Matt Pocock gave a good talk about how he approaches it: https://www.youtube.com/watch?v=-QFHIoCo-Ko

View on HN · Topics

This hasn’t been my experience. State of the art models available to the public still do all sorts of bandaids and bad hacks. Putting code where it doesn’t belong. Stapling types onto variable (in TypeScript) when abstractions/types already exist to use. I use it to generate code, but still have to review every line and have corrections/steering basically every time.

Maybe you have access to some other model?

View on HN · Topics

> To drive a PR to completion, we instruct Codex to review its own changes locally, request additional specific agent reviews both locally and in the cloud, respond to any human or agent given feedback, and iterate in a loop until all agent reviewers are satisfied (effectively this is a Ralph Wiggum Loop ).

https://ghuntley.com/loop/

View on HN · Topics

ok, but you had 1x token to generate , then more 1x to review locally, 1x for the agent local, 1 x for the cloud. then ???x until all bots are satisfied.

You end-up spending at least 5x the amount of tokens for maybe prediction machine to find a discontinuity?

I would say a way better approach is 1.123x to generate code + tests + passing analysis tools + human review + 1x "simplify as much as possible", than letting the snake its own tail without boundaries.