AI-Based Seed Quality Assessment & Farmer Feedback System
B.Tech final-year project using Next.js, Tailwind CSS, Shadcn/UI, Supabase, ResNet50, and Azure AI Foundry with GPT-4o mini. Trained a deep learning model to score seed quality from 0–100 and classify risk as Genuine, Counterfeit, or Suspicious. Developed an end-to-end web application with a farmer complaint module, regional reporting, district-level risk visibility, and contextual AI advisory responses. Solely responsible for the AI model training pipeline and UI/UX design.
MTU Technical Fest Website – FENOMENON 2026
Designed and shipped the official responsive event website using React.js, Next.js, and Tailwind CSS, including event schedules, registration information, and real-time updates for more than 500 attendees.
BoardroomAI Lite – Multi-Agent Startup Strategy Simulator
Built an AI-powered multi-agent system using n8n workflow orchestration. Role-based CEO, CTO, Marketing, and Finance agents collaboratively analyze startup ideas and generate business strategies, MVP plans, budget breakdowns, and investor-focused reports through automated sequential workflows.