Applied data science
Graduate · Data science
Syllabus focus
Topics typically covered
Standard syllabus
Advanced workflow
- Problem framing for ambiguous stakeholder goals
- Data contracts and metric definitions
- Complex wrangling across heterogeneous sources
- Causal vs predictive analysis planning
- Uncertainty quantification beyond point estimates
- Reproducible research at team scale
Experimentation and inference
- Experimental design for online and offline studies
- CUPED and variance reduction survey
- Sequential testing and peeking controls (intro)
- Observational causal methods survey (matching, DiD intro)
- Hierarchical models awareness
- Decision analysis under uncertainty
Modeling at scale
- Feature platforms and training/serving skew
- Model selection under business constraints
- Ensemble and stacking survey
- Calibration and decision thresholds
- Cost-sensitive and constrained optimization intro
- Failure modes and monitoring plans
STEM / applied
Delivery
- Partnering with engineering on pipelines
- Metrics layers and semantic models (survey)
- Stakeholder memos and technical appendices
- Code review standards for analysis PRs
- Postmortems for bad launches
- Capstone-style applied case with defense
Systems awareness
- Warehouse/lakehouse query patterns
- Orchestration of scheduled jobs (Airflow survey)
- Privacy-preserving analytics patterns
- Multi-touch attribution caution (survey)
- LLM-assisted analysis risks and verification
- Portfolio-quality case write-ups
Study units
Each unit includes a study guide, worksheets, review, practice test, and answer key. One unit is free; subscribe for the full class.
- Advanced workflow
Problem framing for ambiguous stakeholder goals
Coming soon - Experimentation and inference
Experimental design for online and offline studies
Coming soon - Modeling at scale
Feature platforms and training/serving skew
Coming soon - Delivery
Partnering with engineering on pipelines
Coming soon - Systems awareness
Warehouse/lakehouse query patterns
Coming soon
Notes
Assumes undergrad DS/stats background. Depth and tools vary by program (industry analytics vs research-prep).