Parth Sangani

Reliable plate extraction in noisy real-world imagery requires more than detection alone, especially under varying lighting. Engineered a four-stage computer vision pipeline using noise filtering, edge profiling, and adaptive thresholding to isolate number-plate regions, then combined YOLOv8n detection and bounding-box analysis with EasyOCR recognition.

ImpactLogged more than 1,000 plate entries to a Supabase (PostgreSQL) schema with confidence scores, creating a strong base for high-volume testing and validation.

Team chat products need low-friction communication, synchronized messaging, and fast routing to feel truly usable day to day. Built a full-stack real-time messaging platform across 5 screens with server-based channels, member directories, and sub-second synchronized delivery over Socket.IO WebSockets, backed by a secure Node.js and MySQL backend handling authentication, sessions, and message persistence across 6 normalized tables.

ImpactDelivered a responsive React front end with client-side routing and reliable sub-second message delivery across channels.

Students learn history better through dialogue than static text, but building a responsive conversational tutor that also measures comprehension is hard. Engineered an AI learning tool enabling real-time conversational interaction with historical personalities via FastAPI and WebSocket services, with adaptive question generation driven by NLP and LLM integration.

ImpactRecognized in the Top 5 among 700+ competing teams at Smart India Hackathon 2024, with adaptive question generation reaching over 85% accuracy in evaluating user comprehension.

Museum staff needed faster visitor support across navigation, ticketing, and contextual queries without scaling manual communication. Built a FastAPI and WebSocket-based assistant with RAG workflows, Supabase integration, and automated email communication.

ImpactReduced manual communication effort by 40% while making visitor guidance faster and more context-aware.

Travel logging apps often feel fragmented across authentication, media storage, and trip management, which makes everyday usage unreliable. Built a polished multi-screen mobile workflow with Firebase Authentication, Supabase Storage, REST-based weather data, and smooth screen transitions.

ImpactImproved media upload reliability by 95% and reduced upload failures to under 2%, making the travel journaling flow significantly more dependable.