What you should be able to do
Choose sensible baselines
Prevent leakage
Evaluate beyond accuracy
Explain model trade-offs
Every module is now a full lesson.
Regression
Predict continuous outcomes using simple baselines before complex models.
Classification
Estimate class labels or probabilities and choose thresholds from real error costs.
Trees
Use recursive feature splits to model nonlinear interactions.
Ensembles
Combine multiple weak or diverse learners to reduce error.
Clustering
Discover groups without labels while validating whether the groups are useful.
Dimensionality reduction
Build a working mental model of Dimensionality reduction then connect it to implementation evaluation and production trade-offs.
Calibration
Build a working mental model of Calibration then connect it to implementation evaluation and production trade-offs.
Projects that prove the skill
Churn intelligence
Use a baseline first then instrument the result and document failure cases.
design with cortex →Fraud baseline
Use a baseline first then instrument the result and document failure cases.
design with cortex →Segmentation study
Use a baseline first then instrument the result and document failure cases.
design with cortex →How progress is judged
Model selection
Evidence should come from code, results, explanation quality and the ability to identify when an approach should not be used.
Metric choice
Evidence should come from code, results, explanation quality and the ability to identify when an approach should not be used.
Failure analysis
Evidence should come from code, results, explanation quality and the ability to identify when an approach should not be used.
Turn the track into a personal plan.
Cortex can break this path into daily sessions and adjust based on completed lessons, interview results and your saved learning goals.
Ask Cortex to plan it →