I kept running into the same problem building AI agents: they forget users, search the wrong docs, hallucinate, and get very expensive as context grows.
Whisper is a context and memory API that gives agents long-term memory, hybrid search (vector + keyword), and delta compression to reduce token usage in multi-turn conversations.
It’s early, but it’s live and I’m onboarding a few developers now. I’d love feedback on whether this solves a real problem for you or what I should build next.