Knowledge BaseConceptsSimulationAI Simulate (Experimental)Sweeper (Experimental)

Sweeper Quickstart

Run a backend-neutral sweep with an injected replay runtime

View as Markdown

Experimental. Sweeper is intended for evaluation and feedback, not production capacity planning.

Install AI Simulate:

$python -m pip install "aisimulate==0.1.0.dev1"

Sweeper requires a RunnerFactory supplied by the application that owns replay execution:

1from aisimulate.sweeper import SmartSearchConfig, Sweeper
2
3config = SmartSearchConfig.from_yaml("sweep.yaml")
4sweeper = Sweeper(runner_factory=my_runner_factory, show_progress=False)
5candidates = sweeper.run(config)

Set sweep.parallel_evals above one to use spawned worker processes. Scripts using that mode must guard their entrypoint with if __name__ == "__main__":.

Next, read the Tutorial for the complete configuration flow or Sweep Configuration Providers to add feature-specific search dimensions.