Tends to be a problem. I've tried to mitigate these problems by using either external harnesses (aka GitHub actions that are "fixed" based on known-good) or by using n-number of witness agents (e.g. Kimi/Qwen/whatever <=> Claude/OpenAI/Google). Generally sucks more time and energy (and now token/$).
that being said, I still have a "fix the code, not the test" line somewhere in here...
8 billion tokens? What did that cost?
No additional cost occured, but just weekly limit exhausted
believing that oneshot prompt is enough specification is pretty delusional.
I keep seeing people talk about the power of these SOTA models, yet keep reading the types of prompts that make no sense to anyone who understands the ludicrous number of decisions that would need to be made.
Yah it’s a weird thing to one prompt. I’m porting some code (Python to go in my case), and I’m pretty specific about doing it sessions by session. Port one file, hit a new wall, break into another session to fix it, and return to my original chat with the problem fixed.