Hi, I'm Satyajeet ! I build reproducible machine learning and analytics projects that translate business problems into measurable models and clear decisions.
Core focus: ML modeling β’ SQL analytics β’ forecasting β’ evaluation & experimentation
Tools: Python β’ SQL β’ pandas β’ scikit-learn β’ matplotlib β’ Git/GitHub (plus Power BI for dashboards)
Projects are designed to show end-to-end capability: problem framing β data β model β evaluation β decision output.
-
E2E ML Pipeline (
e2e-ml-pipeline)
Production-oriented template: preprocessing + training + validation + artifact saving + inference script. -
Time-Series Forecasting (
time-series-forecasting)
Backtesting framework with baselines, feature engineering, and error diagnostics (MAE/RMSE/MAPE). -
Classification Modeling (
classification-modeling)
Strong evaluation: ROC/PR curves, calibration, threshold selection, and interpretability basics. -
SQL Analytics Casebook (
sql-analytics-casebook)
Practical analytics queries: cohorts, funnels, retention, window functions, and performance patterns.
β‘οΈ Portfolio index: ds-portfolio
- Clear README (problem β approach β metrics β results β next steps)
- Reproducible structure (
src/,notebooks/,reports/) - Emphasis on evaluation and decision-making, not only visuals
data/ # sample or synthetic data only
notebooks/ # exploration and experiments
src/ # reusable code (features, pipelines, utils)
reports/ # results, charts, short write-ups
tests/ # lightweight unit tests (when applicable)
README.md
requirements.txt
- β‘ Fun fact: ...
-->