Protocol
End-to-End ML with a Defensible Decision
Choose a real decision problem and deliver a complete ML artifact. A justified recommendation not to deploy is a passing result; unsupported accuracy claims are not.
60–100 active hours60–90 min sessionsAdvanced
Mastery contract
Core protocol · Standard 1.0 · Choose a real decision problem and deliver a complete ML artifact. A justified recommendation not to deploy is a passing result; unsupported accuracy claims are not.
Loading your learning records…
Reasoning — not yet passed
The specification makes ambiguous modeling choices explicit and identifies a non-ML alternative.
Reliable implementation — not yet passed
A reviewer can rebuild features, fit the baseline and reproduce a saved evaluation from the recorded environment.
Reproducible experiment — not yet passed
All model comparisons, 2 ablations and operational tests use the declared budgets and preserve the untouched final holdout.
Defense and handoff — not yet passed
Defend a deployment or no-deployment recommendation in a 6–10 page report with evidence, limitations, rollback and unresolved risks.
All four criteria must pass. Activity completion does not satisfy them.
An independent reviewer reruns your work and varies something you did not rehearse: a fresh input, a different working directory, or a declared edge case.
Pilot decisions are provisional, not external credentials. For an appeal or sensitive artifact, contact your designated pilot operator with the submission ID. Export your assessment history.
What you will do
6 total
Write a 2-page project specification with target population, decision cost, label availability and a falsifiable hypothesis; justify whether ML is needed.
Build a versioned data-to-prediction pipeline with one simple baseline and two justified alternatives; include a fresh-environment reproduction command.
Preregister a fixed split and tuning budget; compare candidates, run 2 ablations, measure uncertainty and test a held-out deployment-like batch.
Debug an injected schema change and one deliberately corrupted label subset; show how the checks and error analysis locate them.
Write a 500–900 word technical report linking the reasoning, code, measurements, and limitations; include the project decision below.
Assess all 4 mastery criteria against saved artifacts, request an independent review, and repeat each failed criterion on a fresh example.
Essentials
A reading session alone never satisfies a mastery criterion.
Store source references, assumptions, and artifact paths beside every result.
Keep an untouched check case that differs from the worked example.
Record all attempts, including failures and results that contradict your prediction.
Use the stated comparison conditions; document every deviation before drawing a conclusion.
Protocol guardrails
Do
- +State the expected result before running the comparison.
- +Keep one minimal reproducible failing case when debugging.
- +Record environment versions and the exact command used.
- +Compare explanations with saved intermediate values.
- +Ask a reviewer to challenge the weakest assumption.
Don't
- ×Do not copy a worked solution and present it as an independent implementation.
- ×Do not tune against held-out evaluation outcomes.
- ×Do not report only the best seed or discard inconvenient runs.
- ×Do not equate elapsed hours or a completed run with a passed assessment.
- ×Do not conceal reduced-scale experiments behind claims about the original full-scale result.
Protocol authorship
Written by NuthinButta as instructional design. The teaching sources are the official references linked in each milestone; the exercises, workload and pass thresholds are ours, not their authors'.