A smart AI chatbot built for Keshav Memorial Institute of Technology (KMIT) to help students, faculty, and visitors instantly access important information, events, and resources via a floating assistant on the college website.
- 💬 Conversational AI powered by Gemini LLM (Google Generative AI)
- 📚 Retrieval-Augmented Generation (RAG) using FAISS + Sentence Transformers
- 🗃️ College info and event data loaded via JSON files and Supabase
- 🔗 Quick redirects to portals like Attendance, Exam Info, Syllabus, etc.
- 🧑🎓 Smart responses to FAQs, context-based queries, and navigation requests
- 🖼️ Floating chatbot UI embedded on the website using Vite + TypeScript
- 🔐 CORS-enabled Flask backend serving AI and database responses
| Layer | Tech Used |
|---|---|
| 🧠 AI Model | Gemini API (Google Generative AI) |
| 🔍 Embeddings | Sentence-Transformers (MiniLM-L6-v2) |
| 📥 Vector DB | FAISS (Facebook AI Similarity Search) |
| 🌐 Backend | Flask + Flask-CORS + Supabase SDK |
| 📊 Data | JSON file + Supabase DB |
| 🧪 Scraping | Selenium (ChromeDriver) |
| 🖥️ Frontend | Vite + TypeScript |
- User submits a question via the floating chatbot.
- Query is normalized and embedded using Sentence Transformers.
- FAISS retrieves the most relevant context from the JSON data.
- Context and query are sent to Gemini API for the final response generation.
- Supabase is queried for dynamic content like event updates.
- Response is displayed in the chatbot UI.
For questions or to see a demo, feel free to reach out!