Orchid – Local-first record and replay for AI agent debugging
Deterministic replay of agent runs without mocking—that's genuinely new.
Record, replay, and improve AI agents in production, built on ZenML
Checkpoint replay and what-if experiments compete directly with LangSmith.
AI engineers building production agent systems
LangSmith · Arize Phoenix · Helicone
It's OSS and free to use. Would love some feedback from the community.
Deterministic replay of agent runs without mocking—that's genuinely new.
Runtime clicks for you instead of waiting for user input like shepherd.js.
Flight recorder for AI agents: record, replay, enforce policies on every LLM call.
Record production Python bugs and step backwards from crash to cause in VS Code.
Conceptual art piece masquerading as infrastructure with one registered agent.
Iteratively improves agent harnesses from 67% to 87% on tau-bench using production traces.