InterviewsVector
Capability spine

Arc 03 · AI Engineer

Learning Systems and Classical ML

Build the discipline that still governs foundation-model products: splits, baselines, leakage control, calibration, drift, and causal humility.

Exit capability: Ship and defend a supervised learning system with trustworthy evaluation.

Mapped lessons
7
Published now
0
Full-arc estimate
15 hours
Last edited
2026-08-11

Lesson sequence

Live units open into complete labs. Planned units stay visible to show the dependency path, but intentionally have no detail route.

  1. 01

    The Generalization Contract

    Planned

    Define the population, task, loss, and deployment distribution a model is expected to generalize to.

    Concept labFoundation80 min estimate

  2. 02

    Split Data by Causality, Not Convenience

    Planned

    Use time, identity, groups, and provenance to prevent leakage that random splits conceal.

    Failure labIntermediate95 min estimate

  3. 03

    Linear Models as Debugging Instruments

    Planned

    Use linear and logistic regression to expose data and target problems before adding model complexity.

    Build labIntermediate100 min estimate

  4. 04

    Trees, Boosting, and the Shape of Residual Error

    Planned

    Understand how tree ensembles partition mistakes and when their inductive bias wins.

    Build labIntermediate110 min estimate

  5. 05

    Calibration, Thresholds, and Decision Cost

    Planned

    Turn scores into actions using reliability curves and asymmetric error costs.

    Systems labIntermediate100 min estimate

  6. 06

    When the Product Needs Ranking, Not Classification

    Planned

    Choose ranking losses and metrics when ordering quality matters more than labels.

    Concept labIntermediate90 min estimate

  7. 07

    Drift, Feedback Loops, and Delayed Labels

    Planned

    Design monitoring when the system changes the data it later learns from.

    Systems labAdvanced110 min estimate