Developed a full-stack chatbot for KL University using HTML, CSS and JavaScript for the frontend and Node.js + Express.js for the backend, integrating Google Gemini, Groq, DeepSeek, Qwen and other AI models to answer admission and campus queries through real-time web scraping, with voice interaction using ElevenLabs TTS, Vapi and Tavily Search API for 24/7 AI-assisted student support.
• Main Challenge in the project: Keeping answers accurate and consistent while juggling multiple AI models was the biggest challenge — every model behaves differently, so routing each query to the right model, unifying the responses, and keeping real-time scraped campus data fresh without slow replies took the most work. Making voice interaction feel instant despite speech-to-text and TTS latency was equally tricky.