Data science capstone
Undergraduate · Data science
Syllabus focus
Topics typically covered
Standard syllabus
Project scoping
- Choosing a feasible question and success metrics
- Data access, licenses, and IRB/privacy checks
- Risk register: schedule, data quality, model risk
- Milestone plan from proposal to final demo
- Advisor/stakeholder check-ins
- Defining out-of-scope clearly
Execution
- End-to-end pipeline: ingest → clean → explore → model/eval
- Reproducible environments and seeded runs
- Version control and experiment notes
- Validation strategy appropriate to the claim
- Ablations and sensitivity checks
- Backup plans when data or APIs fail
Delivery
- Written report structure for technical and non-technical readers
- Figures/tables that carry the argument
- Live demo or recorded walkthrough
- Limitations, ethics, and next steps
- Code/README handoff quality
- Peer review and revision cycles
Professional practice
- Timeboxing and scope cuts under deadline
- Collaborating in pairs/teams when required
- Responding to critical feedback
- Citing data and software correctly
- Archiving artifacts for grading and portfolio
- Reflecting on skills gained and gaps
Notes
Standard-track integrative course. Tooling follows prior DS sequence (Python/R/SQL). Tutoring focuses on scoping, reproducibility, and communication under the course rubric.