A recap should help you do the next small thing.
Start with the system boundary
The first session defines which component is being measured before comparing tools or approaches.
Watch the failure walk-through
The second begins with logs and constraints, not a polished success story or an unsupported benchmark.
Try the smallest useful case
The final card provides prerequisites, expected output, and a stop condition for a local test.
