Why we are building LogicMem in public
LogicMem has been running our own agents every day. Claude, GPT-based agents, Cursor and local models all read and write the same memory, and every write lands in a hash-chained audit log that now holds about 800,000 entries that verify.
What we do not have yet is outside users. Until someone else runs it, every claim is just a claim. So we are doing two things:
- Building in public. A weekly note on what our agents remembered, contradicted and caught, with real numbers.
- Recruiting 5 to 10 design partners who run agents from more than one vendor and want them to share memory.
Why this angle? Memory for a single assistant is now free in ChatGPT, Claude and Gemini. What is still missing is one live memory shared across vendors, with receipts. That is the gap we are going after.
If that is you, apply here.