Projects

A curated portfolio of my projects in data science, big data, machine learning, deep learning, and MLOps.

Production RAG System with Hybrid Retrieval, Citation Enforcement, and CI-Gated Evaluation

A production RAG pipeline featuring BM25+vector hybrid retrieval, cross-encoder reranking, enforced source citations with decline-to-answer behavior, and a three-vendor…

Customer Care ETL Pipeline

A simple extract-transform-load (ETL) pipeline that helps streamline customer care and support data.

Taxi Demand Prediction

An application to forecast the demand of yellow taxis in New York City.

YouTube Comment Intelligence

An end-to-end MLOps system that delivers real-time sentiment analysis of YouTube comments through a Chrome extension, backed by a FastAPI service deployed on AWS.

Food Delivery Time Prediction

A machine learning project that predicts how long a food delivery takes.

Olist E-Commerce Database

Designed and optimized a normalized e-commerce relational database (PostgreSQL → SQLite) using the Olist dataset, with BCNF validation, indexing, query performance analysis…

Spotify Hybrid Song Recommender

A song recommender system built using a hybrid approach of content-based filtering and collaborative filtering.