Reasoning infrastructure2026Active
Smriti
Checkpointed reasoning state for long-running, multi-model work.
A narrow list of the independent systems I am actively building.
The independent systems I am actively building right now.
Checkpointed reasoning state for long-running, multi-model work.
A local-first trust kernel for agent systems with policy checks, append-only history, and verification.
A property-diligence system that gathers fragmented records, surfaces inconsistencies, and keeps judgment with the human operator.