SupportIQ - Support Chatbot For Websites

An intelligent, autonomous customer support platform powered by Google's Gemini LLM. SupportIQ automatically crawls and indexes company documentation to provide accurate, context-aware answers to user queries in real-time.

SupportIQ Stack: Next.js, Google Gemini API, Supabase, Firecrawl Description: A full-stack RAG (Retrieval-Augmented Generation) application designed for automated customer success.

• Data Pipeline: Utilized Firecrawl to scrape and turn web documentation into clean markdown.
• Vector Search: Implemented Supabase for storing vector embeddings, allowing the AI to semantic search through company data for precise context.
• Generative AI: Integrated the Gemini API to synthesize natural, human-like responses based on the retrieved context.
• Frontend: Built a responsive chat interface with Next.js, featuring real-time streaming responses and a custom backend.

The point of SupportIQ is not "a chatbot." It is a retrieval loop you can defend in a standup: crawl the docs, chunk them, retrieve the passage, generate the answer, and show the source. If the model cannot cite the page, it should say so.

Live: supportiq-ten.vercel.app. Source: github.com/xheenkhalil/supportiq.