An AI app can keep running after a prompt edit, model swap, or dependency update while its answers get worse.
We built EvalCore to catch those changes in CI with repeatable evals and offline replay.
Built with @manyamabhishek .
Live on Product Hunt:
https://t.co/BXegybtGkk
the annoying part about dots isn't privacy, it's that memory is all or nothing. if it learns something wrong about you, you can't remove that one thing, you delete the whole dot and start over
you can tell your dot to "forget" something but per their faq you can't actually delete individual memories. so it's not forgetting, it's just choosing not to bring it up