About
I'm building HUQAN, a local-first trust layer that judges AI agent actions before they execute — file writes, API calls, database changes — and produces deterministic allow / review / block decisions with an auditable Trust Receipt for every call. Solo founder, engineering-heavy, evidence-obsessed: 1,900+ tests, deterministic replay, no cloud dependency. Currently in the İTÜ Çekirdek incubator (Istanbul), running toward first pilots. Open source core: github.com/huqan-com/huqan DMs open. Also building Levh, a local-first memory layer for AI agents — born out of my own frustration with losing context while juggling multiple agents day to day. Launching soon, still learning as I go. Happy to hear feedback or criticism.
Badges




Forums
View more
