‹ BackHN Continuity

Thread

I don't want to read what you didn't write

1070 points · 460 comments · mooreds

  1. caarmen · · focus · HN ↗
    To address the problem of symmetry, where the effort to generate a wall of LLM text is much lower than the effort to review it: if someone throws such a wall of text to me, I suggest we review it together. Same for huge PRs. When a coworker opens a PR with thousands of lines of non trivial code, I schedule a meet with them to do a pair review together. When doing this for a PR review, I found it much easier for me to review, and the other person also caught mistakes in their PR that I might have missed. This pair review (of PRs or LLM generated text) adds a tiny bit of humanity to the process.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.