Scenarios
One of the powerful features of BotDojo is its ability to view how evaluations run with different models, prompts, and indexing strategies. By leveraging the evaluation framework, you can gain valuable insights into the behavior and effectiveness of various configurations.
You can test different scenarios by modifying parameters in your flow, but to do this well at scale, you should build a DataSet of tests and run these tests in bulk using Batches. Batches allow you to override properties on each run, where each evaluation represents a unique combination of model, prompt, and indexing strategy.
