experiment

An experiment run produces the smallest valid amount of evidence for a research quest by automating execution. It applies to quests requiring precise measurements, baselines, selected ideas, and explicit evaluation contracts.

The best way to validate research results is by running main experimental runs through a structured workflow with strict validation and error handling. This ensures the production of trustworthy, verifiable evidence for your baselines and selected ideas.

Yes, you can use this to run deterministic tasks for precise measurements and comparisons. It executes main experimental runs that adhere to structured workflows and explicit evaluation contracts to produce verifiable evidence.

You should not use automated experiment runs if your research quest lacks an accepted baseline, a selected idea, or explicit evaluation contracts. The workflow requires strict validation and structured inputs to produce trustworthy evidence.