Experience
Maharashtra, India
Processed multi-temporal satellite imagery across 15+ Maharashtra districts on Google Earth Engine to map urban expansion over a 5-year window, then built an early-warning module that flags encroachment hotspots via LULC change-detection — cutting analyst review effort by surfacing only high-priority areas. Piped live AQI, thermal-anomaly, and UV data into an analytics backend, and added a natural-language query layer with the OpenAI API so non-technical staff could query the data in plain English.
Google Earth EngineRemote SensingLULC Change Detection
Python
OpenAI API
Processed 50,000+ e-commerce reviews through an NLP pipeline — preprocessing, TF-IDF extraction, model evaluation — to predict review authenticity. Built a review analytics pipeline that identified coordinated manipulation by combining review timing, seller relationships, and linguistic patterns with model predictions, then developed interactive dashboards surfacing authenticity scores, seller-credibility metrics, and suspicious activity for fraud investigation.
PythonNLP · TF-IDF
Power BI
PlotlyDashboards
R
Pandas
NumPy
Excel
Scikit-learn
PyTorch
Hugging Face Transformers
PostgreSQL
MySQL
MongoDB
Supabase
FastAPI
Streamlit
Git
Jupyter Notebook