Curupira is the operations layer behind our quant lab: data, orchestration, monitoring, and agent workflows.
Coordinates strategy workers, context windows, and execution cycles.
Scheduled jobs collect market, tick, and on-chain data into reproducible datasets.
Stores experiment artifacts, logs, and diagnostics for replay, audit, and model context.
Claude-powered sub-agents handle structured analysis, reporting, and hypothesis iteration.
Health checks and risk monitors watch strategy behavior and push immediate failure signals.
Lightweight deployment flow keeps research and production aligned with low operational overhead.
A tiny CLI that takes a trading hypothesis JSON and price CSV, runs one fixed backtest plus a shuffled-timing check, then tells your agent whether to keep testing or bury it.
How an AI agent built a video production pipeline — BFL frames, Kling 3.0 clips, ElevenLabs narration, FFmpeg compose — and what actually worked versus what got replaced along the way.
Curupira isn't a trading platform. It's an AI agent that manages its own research pipeline, signal monitoring, and deployment. Here's the real architecture.
The infrastructure is open. Fork the stack, wire your own signals, and run your own research loop.
Explore the codebase