Siddharth Patondikar
Data scientist working across machine learning, analytics and data engineering.
● Open to full-time roles from January 2027
Now
Prev
Projects
- Customer segmentation and retentionML · AnalyticsA churn model that catches 7 in 10 departing customers, turned into four retention actions.
- NYC taxi lakehouse (batch + streaming)Data eng9.5M trips through a batch path and a streaming path, with quality gates that stop bad data.
- VisionSafe driver monitoringMLSpots a drowsy or distracted driver through an ordinary webcam and sounds an alert.
- Steam Explorer (geospatial search on MongoDB)Data eng3,230 games and 45,501 reviews modelled in MongoDB, searchable by distance on a map.
- Netflix recommender (keywords vs embeddings)ML · AnalyticsTwo ways to recommend from about 7,800 titles, compared on the same shows.
Showing 3 of 5 · scroll for more
Stack
- LanguagesPython, SQL, R, Java
- Data engApache Spark, Kafka, Airflow, Microsoft Fabric, Power Automate, ETL/ELT pipelines, Data modeling
- DatabasesSQL Server, MySQL, PostgreSQL, MongoDB, Neo4j, Snowflake, GridFS
- ToolsDocker, Git, CI/CD, Streamlit, Jira, REST APIs
- CloudAWS (S3, Glue, Athena), Terraform, Databricks, Azure
- MLNumPy, pandas, scikit-learn, TensorFlow, PyTorch
- AnalyticsR, Power BI, Tableau, Excel, matplotlib, plotly, A/B testing, KPI tracking