Evaluate a design

Choose evidence that answers a concrete decision.

Questions to answer

What is held constant? What is the independent oracle? What result would disprove my claim?

Start here

Read methodology, fairness controls, and findings. Run python -m arena validate and a vanilla mock arena before designing a new comparison.

Prerequisites: a small task you can describe, its permitted effects, and basic Python for executable examples. Reading the guides needs no API key.

Evidence and coverage

Deterministic mechanics are available. Model quality needs repeated native-provider trials; full-harness and framework-only results must remain separate.

Follow the concept and build path for runnable lessons and the ecosystem map for wider coverage. Return to all journeys.