Cluster 05
Experimenting with AI
Design and run experiments to test whether AI actually improves specific steps in your workflow.
5 Lessons
01
Designing an Experiment
Framing a clear hypothesis: AI can improve step X by doing Y.
02
Choosing the Right Model
When to use different models and how to decide.
03
Running and Comparing
Using A/B testing to compare models and prompt variations side by side.
04
Evaluating Outputs
How to rate results honestly using the 1-5 scale and why your notes matter.
05
Iterating on Prompts
The cycle of test, evaluate, refine â and knowing when to stop.