Machine Learning
Milestone 2 of 7 · industry roleplay

Baseline, splits & the right metric

Split the data the way it will be used, build your first models (a baseline, logistic regression, then a proper preprocessing pipeline), forecast with linear regression and gradient descent, regularise with ridge and lasso, fix a broken k-NN, and only then evaluate: agree the metric with the clinic, plot the results and choose a cost-based threshold. Ends with communication, code review and debugging. Includes 7 hands-on Python coding tasks (pandas, scikit-learn), checked by automated tests.

Rigour50Communication50Ethics50Impact50Teamwork50
Mon 9:15Sprint planningMon 14:00Decide the splitTue 10:00Code it: build your first modelTue 15:00First resultsTue 16:30Code it: an honest baseline pipelineWed 10:00Forecast the weekWed 14:00Code it: gradient descent from scratchThu 9:30Too many featuresThu 11:00Code it: Lasso with cross-validationThu 15:00Why is k-NN so bad?Wk 2 · Mon 10:00Agree the metricWk 2 · Mon 14:00Code it: the evaluation chartsWk 2 · Tue 10:00Choose the thresholdWk 2 · Tue 13:00Code it: the threshold from costsWk 2 · Wed 9:05"Can we say 95% accurate?"Wk 2 · Wed 12:30Post the weekly updateWk 2 · Thu 10:00Code reviewWk 2 · Thu 16:30Code it: fix the broken nightly jobWk 2 · Fri 11:00Track the experiment
Mon 9:15Sprint planningStand-up
Tomás
Maya
Tomás RiveraSenior ML Engineer · your mentor
New

Bluegum Health, Riverside Clinics and every person here are fictional. The decisions, tools and practices are modelled on how real ML teams work.

Ready