cool episode from the @changelog (changelog.com/podcast/676)

I recommend listening to it! I think I am spotting some circular logic: "you" are building software with "agents" but you can't trust them so you build tests suite, QA process and more (also using LLM)... Not sure you can solve trust like that.

Similar for code review, you need to review more but you can´t build mental model while writing code.

Interesting take also at the end on FOSS (GNU considered?).

2026 year or trust?

0

If you have a fediverse account, you can quote this note from your own instance. Search https://fosstodon.org/users/defuneste/statuses/116054595329782077 on your instance and quote it. (Note that quoting is not supported in Mastodon.)