Learning Path

Path A: New to Applied ML

1. projects/sota-supervised-learning-showcase
2. projects/sota-unsupervised-semisup-showcase
3. projects/causalml-kaggle-showcase

Path B: Decision Science / Causal Focus

1. projects/causalml-kaggle-showcase
2. projects/xai-fairness-audit-showcase
3. projects/mlops-drift-production-showcase

Path C: ML in Production Focus

1. projects/sota-supervised-learning-showcase
2. projects/mlops-drift-production-showcase
3. projects/batch-vs-stream-ml-systems-showcase
4. projects/model-release-rollout-showcase

Path D: Modeling Optimization Focus

1. projects/sota-supervised-learning-showcase
2. projects/automl-hpo-showcase
3. projects/rl-bandits-policy-showcase

Path E: Feature and Representation Focus

1. projects/sota-supervised-learning-showcase
2. projects/feature-engineering-dimred-showcase
3. projects/sota-unsupervised-semisup-showcase

Path F: Short Course (Two Weeks)

• Day 1-2: sota-supervised-learning-showcase
• Day 3-4: feature-engineering-dimred-showcase
• Day 5-6: automl-hpo-showcase
• Day 7-8: xai-fairness-audit-showcase
• Day 9-10: mlops-drift-production-showcase
• Day 11-12: batch-vs-stream-ml-systems-showcase
• Day 13-14: rl-bandits-policy-showcase or causalml-kaggle-showcase

Path G: Contract-First Supervised Workflow

1. projects/eda-leakage-profiling-showcase
2. projects/feature-engineering-dimred-showcase
3. projects/automl-hpo-showcase
4. projects/xai-fairness-audit-showcase

Path H: Credit Risk Capstone Workflow

1. projects/eda-leakage-profiling-showcase
2. projects/feature-engineering-dimred-showcase
3. projects/credit-risk-classification-capstone-showcase
4. projects/xai-fairness-audit-showcase

Path I: Ranking and Serving Workflow

1. projects/learning-to-rank-foundations-showcase
2. projects/ranking-api-productization-showcase
3. projects/model-release-rollout-showcase

Path J: Forecasting and Observability Workflow

1. projects/nyc-demand-forecasting-foundations-showcase
2. projects/demand-api-observability-showcase
3. projects/mlops-drift-production-showcase

How To Know You Are Progressing

• You can explain outputs in plain language.
• You can justify model choices with evidence.
• You can describe one limitation or risk per method.
• You can propose a production or governance guardrail for each modeling workflow.

Coverage Cross-Reference

Use docs/aspect-coverage-matrix.md to confirm which project demonstrates each method (splits, imbalance handling, explainability, HPO, tracking, and productionization).

Track Pages

For track-level documentation with artifact-focused guidance:

• docs/tracks/foundations.md
• docs/tracks/production.md
• docs/tracks/ranking.md
• docs/tracks/forecasting.md
• docs/tracks/responsible-ai.md
• docs/tracks/optimization.md