The pre-LLM reflex layer for AI agents. library with a live training server, microsecond latency, learns from LLM corrections — System 1 to your LLM's System 2.
intent triage prefilter intent-classification intent-detection multi-intent microsecond-resolution llm tool-routing
-
Updated
Apr 28, 2026 - Rust