jasbury

Because that model does not scale for teams of developers who need a consistent view of CI health, build health, and PR readiness

> Chatting through all the races and the different candidates, he worked to figure out who best represented his values.

This sounds like it can be a really good thing. In an ideal world it’s a perfect use of LLMs. It helps make the ballot more approachable to people, and it can encourage more people to vote or help people vote less blindly.

At the same time, it also relies on these tools being unbiased adjudicators. And over time I think there’s a greater risk of this straying from the truth.

Well if the sentry chip has its own sentry chip, things can rarely ever go wrong! Am I right?

The author is really just advocating for design docs, which do make sense in a repo. LLMs are excellent at creating .md docs tracking design decisions, code architecture, trade-offs, etc. And these actually can be of great use to future contributors.

For me, long-running tasks are not about generating a lot of code. I’m very picky about what my code looks like. But I’ll happily run for long periods of time debugging problems and/or doing testing and validations. Depending on the problem space, this could mean hours of work for each iteration while it attempts to find a working solution

This. I was fed up today with constantly correcting Opus for a specific task. So I finally decided to try Astra. It handled all of my prompts in one go.