Portfolio
All Projects
A comprehensive collection of my work spanning various technologies and domains.
MarketMuse AI
Engineered a comprehensive, multi-agent AI marketing platform using React and Supabase. Architected a sophisticated, real-time token-based billing and usage-tracking system for 10 distinct AI agents. Implemented a full-cycle campaign management dashboard with data-driven analytics and integrated n8n webhooks for complex workflow automation.
ChartGenie
Architected a full-stack, AI-powered data visualization platform to transform raw CSVs into 7+ types of interactive dashboards using React and Apache ECharts. Engineered a high-performance RAG (Retrieval-Augmented Generation) system, leveraging Groq's ultra-fast LPU and Supabase Edge Functions, to enable complex data analysis via a natural language chat interface.
PropertyRAG
Engineered a production-grade real estate intelligence platform powered by Retrieval-Augmented Generation (RAG). Architected a sophisticated backend (FastAPI) with 9-intent classification, a deterministic analytics engine, and a hybrid search pipeline with CrossEncoder re-ranking to deliver high-confidence, multi-turn answers.
Customer Churn Retention
Developed a machine learning model to predict customer churn using telecom industry datasets. Implemented feature engineering, model selection, and evaluation to optimize churn prediction accuracy, aiding customer retention strategies.
Room-wise Object Detection
Developed a computer vision application using YOLOv8 and Streamlit to detect and analyze unique household objects. Implemented a full-featured analytics dashboard and a batch processing module to provide comprehensive object statistics and distribution charts from multiple images.
Distributed Image Processing Pipeline
Designed and implemented a distributed system for large-scale image processing tasks. Utilized distributed computing frameworks to parallelize image operations, enhancing processing speed and scalability for high-volume datasets.