TenseMem is durable agent memory with a deterministic resolver, enforced cardinality, freshness-aware reads, and lasting corrections — so your agent gets what still holds, not a pile of conflicting prefs.
Building TenseMem around a simple failure mode: agents don t forget facts - they forget which ones still apply (prefs flip, corrections get buried, retrieval serves two current truths).
The problem: agents don’t forget facts - they forget which ones still apply. Preferences flip (email → slack → maybe email again), corrections get buried, and retrieval serves a pile of conflicting “current” truths.
TenseMem is durable agent memory that answers what still holds — with a deterministic resolver, enforced cardinality, freshness-aware reads, a contamination guard, and corrections that stick.
Try it
pip install tensemem
SQLite-local by default · Studio UI · open source on GitHub
Who it’s for Builders shipping LLM agents who need memory that resolves, not just stores.
Would love feedback from anyone who’s fought stale prefs or "I already told you that." Upvotes and tough questions both welcome- happy to dig into the design in the thread.
Hey Product Hunt 👋
Maker of TenseMem here.
The problem: agents don’t forget facts - they forget which ones still apply. Preferences flip (email → slack → maybe email again), corrections get buried, and retrieval serves a pile of conflicting “current” truths.
TenseMem is durable agent memory that answers what still holds — with a deterministic resolver, enforced cardinality, freshness-aware reads, a contamination guard, and corrections that stick.
Try it
SQLite-local by default · Studio UI · open source on GitHub
Who it’s for
Builders shipping LLM agents who need memory that resolves, not just stores.
Would love feedback from anyone who’s fought stale prefs or "I already told you that." Upvotes and tough questions both welcome- happy to dig into the design in the thread.
Links:
GitHub: https://github.com/tensemem/tensemem
PyPI: https://pypi.org/project/tensemem