I am building toward Applied / Product Data Science with strong ML Engineering skills. I care about the full path from a product question to a trustworthy decision: define the metric, validate the data, model the uncertainty, test the intervention, and ship a reproducible system.
| Depth I am developing | Breadth I am building | Long-term direction |
|---|---|---|
| Experimentation · Causal inference · Applied ML | SQL · PySpark · MLOps · Cloud · LLM systems | Applied / Product Data Scientist → Applied Scientist / ML Scientist |
I treat this as a roadmap, not a wall of skill badges. A technology appears as a demonstrated strength only after a project makes the design choices, limitations, and evidence visible.
Three case studies are moving through a public-release review. Each repository is rebuilt from my own analysis and excludes private course material, restricted data, and unverifiable claims.
- Conversion Intelligence — decision-aware conversion modeling, leakage-safe pipelines, cost-sensitive thresholds, and an A/B-test handoff.
- Lifecycle Email Experimentation — multi-arm messaging analysis with causal inference, multiplicity control, temporal attribution, and funnel integrity.
- Amazon Review NLP — group-aware sentiment evaluation and a compact, reproducible text-classification pipeline.
Links will be added only when each project passes its data, license, reproducibility, and documentation gate.
| Frame the decision | Separate prediction from causality | Build for review |
|---|---|---|
| Start with the user, metric, prediction time, and cost of error. | Use observational models for ranking; use experiments for intervention claims. | Add data contracts, tests, CI, model cards, and honest limitations. |
- Product conversion modeling and experimentation handoff
- Multi-arm lifecycle messaging experimentation
- NLP evaluation and deployment discipline
- SQL/PySpark feature pipelines, MLOps, AWS, and LLM systems as deeper extensions—not decorative add-ons
Measure carefully · Build responsibly · Improve in public