Maya C.
Toronto · remote
Building
Eval infrastructure for AI agents. Tests for trust, drift, and goal preservation in long-running autonomous systems.
Why you should meet
Maya builds the eval layer for autonomous systems. You build the autonomous systems. Worth comparing how each of you defines trust before drift kills the meeting.
Opener, half-drafted
Saw your eval suite for goal preservation in long-running agents. I am running into drift from the building side, not the measuring side. 30 min to compare notes?