4 projects
agent-regress-cli
Statistical regression testing for LLM agents. Compare two agent versions, get p-values on behavior change.
agent-observability-trace
AI agent observability with deterministic record/replay — capture any LangGraph or OpenAI Agents SDK run, replay it offline at zero API cost.
memtrust
Independent, reproducible benchmark harness for agent-memory backends
ownvoice
Train a LoRA voice adapter for pocket-tts and own the resulting model, no API lock-in.
