Aviraj Sapate

Autonomous RAG Document Q&A System

Engineered an end-to-end Retrieval-Augmented Generation (RAG) pipeline to parse unstructured PDFs and enterprise documentation. Built vector embeddings using ChromaDB and exposed lightweight FastAPI REST endpoints for low-latency querying with precise context bounds.

Customer Churn Prediction Engine

Developed an end-to-end machine learning pipeline to predict customer churn from structured transactional and usage data. Performed data preprocessing, feature engineering, and model evaluation using Scikit-learn, achieving high accuracy and deploying the trained model into a production-ready container.