logoalt Hacker News

arw0n • today at 8:30 PM • 1 reply • view on HN

I have inherited a multi-repo project that is actually really well designed, and I like the clear separation of concerns between components and very solid APIs connecting them. But since they have individual deployments, and could get out of sync, Astra/Sol keep insisting we have huge json contracts generated by unreadable bash scripts, enforced by custom CI jobs, and supported by a huge number of tests.

I've seen two projects where developers didn't curtail this kind of overengineering, and recently had to spend a very significant amount of time and tokens to debug their deployment issues, because even for advanced agents there's a point where each issue could be caused by 12 things. This is especially pronounced in DevOps, but I see it in code as well; the suggestions are all reasonable individually, but you quickly end up with a code base that is too complex for a human to reason about, and it currently only kind of still works with agents because capabilities are growing so quickly.


Replies

jakevoytko • today at 8:46 PM

I gave Astra full control over the design of a software factory for a few days, as a side project. Then I switched to making it justify every change that it wanted to make. It solves problems that don't exist, has tons of random verifications that don't add value, selects tortured architectures instead of asking if it can relax its assumptions a bit, etc.

Typical convo:

Me: "Hey, I see behavior X, which seems like a bug. What is causing that?"

Astra: "That's caused by a verification failure; we should be verifying the hash of this JSON output along with a permanent audit record. Let me know if you'd like to make that change."

::ask it what caused X to happen 6 or 7 times::

Astra: Tool use was limited to 1GB but the Rust compiler needed more than that, and that caused some PROBLEMS

Me: Remove that constraint

Astra: OK