Developed an AI-powered email client with a RAG pipeline for intelligent message prioritization and semantic search. Reduced inbox triage time by 60% through smart reply suggestions, automated categorization, and real-time WebSocket sync. Built a type-safe full-stack API with Prisma ORM and PostgreSQL, implementing row-level security and comprehensive query optimization.

Developed full-stack Point-of-Sale system for restaurants using React, Tailwind CSS, Node.js, Express.js, and MySQL, serving 50+ active clients. Implemented role-based access control (RBAC), inventory management, real-time sales analytics, and automated receipt generation. Containerized application with Docker and deployed on cloud infrastructure with CI/CD pipelines, achieving 99.9% uptime.

Built a vehicle AI assistant with ML-driven recommendations, TSP/VRP optimization, real-time GPS tracking, and live operational dashboards. Developed full-stack architecture using Python, Django, TypeScript, Next.js, and TensorFlow, processing real-time sensor data with sub-100ms latency. Deployed containerized microservices using Docker and Kubernetes with automated health checks, rollback strategies, and zero-downtime deployments.

Production-ready streaming platform for drama, series, and short-form video content. Built with modern web technologies, it provides a seamless user experience for browsing and watching content, while offering administrators powerful tools for content management, AI-assisted generation, analytics, and monetization