One ecosystem. Four products.
The CLI learns from every task. The Router learns from every request. The LLM learns from every trajectory. The SDK exposes it all programmatically. A self-improving flywheel.
Three nodes. One compounding loop.
Three products, one flywheel. The CLI emits trajectories. The Router learns which models win where. The LLM predicts optimal pipelines — which the CLI runs next.
Every task makes the next one better.
Developer uses Brainstorm CLI to build a feature.
CLI routes the task through BrainstormRouter to the optimal model.
Router records the outcome — which model, what cost, did it succeed.
BrainstormLLM trains on accumulated trajectories, improving predictions.
Next time, routing is smarter. The system compounds.
The Stack
Pick a starting point. Gain the whole ecosystem.
Install the CLI and your trajectories feed the Router. Use the Router and its routing sharpens over time. Every surface compounds on the others.