Jev in production › Agent context and memory

Jev Recall (samdotmak)

Python library where Jev answers one calibrated yes/no per stored memory in a single request and keeps memories above a threshold instead of ranking by embedding similarity. Passed 17/18 benchmark requests (author).

Open on GitHub ↗

17/18measured against a baseline, as published by the source
Use
Agent context and memory
Industry
AI infrastructure
Form
Open-source tool
Stage
Beta
Listed
2026-09-20
Found via
github
Repository
samdotmak/jev-recall
Stars
38
Forks
3
Last push
2026-09-20
Language
TypeScript
License
MIT

The README opens with

Jev Recall gives you LLM re-ranker quality retreival at semantic search prices and speed

Given a user query and a pile of memories, jev recalls returns the most relevant memories, judged by TypeSafe's Jev model, one calibrated yes/no per memory, in a single request.

Try the live demo → · replays real runs, or runs live with your own API keys.

With LLMs, writing memories is easy. Knowing which ones to retrieve is the hard part.

- Keyword search works only if you already know what you're looking for. - Semantic search works only if the memory that matters happens to look like the request.

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for agent context and memory