Model Mix Analytics
Model Mix Lab: Scenarios in Python
Translate scenario tables into vectorized calculations you can diff across versions.
Tuition is informational only. Enrollment staff confirm availability and policies by email.
Request a syllabus callOverview
Build transparent scenario engines using plain Python and pandas. You will version assumptions, compare branches, and export narrative-ready tables. The tone stays operational: no black-box optimizers, only inspectable steps.
What you build
- Scenario matrices with explicit drivers
- Deterministic seeds for reproducible runs
- Diff views between scenario branches
- Lightweight visualization of sensitivity bands
- Packaging assumptions into config files
- Docstrings that read like internal memos
- Pair exercises on naming conventions for metrics
Outcomes
- Ship a three-branch scenario pack with documented drivers
- Explain how a change in one driver propagates through dependent rows
- Archive a tagged release others can re-run
Mentor on record
Noah Im
Curriculum analyst translating messy desk requests into teachable modules.
FAQ
Will we use optimization libraries?
No. We stay with linear combinations and piecewise rules you can audit on paper.
Can teams attend together?
Yes. Pair seats receive a shared review channel with the instructor during office hours.
What is out of scope?
We do not teach probabilistic simulation packages or GPU workflows.
Participant notes
The diffable scenario pattern replaced a fragile macro sheet for our quarterly outlook pack.