All projectsLab Project
ASTU Intelligent Knowledge Base (RAG-AI)
A Retrieval-Augmented Generation (RAG) chatbot platform demonstrating enterprise-grade document search and AI-powered Q&A over institutional knowledge bases.
The problem
Institutional knowledge is scattered across documents, making it difficult for students and staff to find accurate information quickly. Traditional search fails to understand context and nuance.
The solution
Built a RAG-based chatbot platform with document ingestion, vector embeddings via pgvector, and grounded generation that cites sources. Includes admin panel for document management and analytics.
Business value
Provides instant, accurate answers with verifiable citations. Reduces support workload and serves as a reusable pattern for client-facing knowledge bases and internal search tools.
Technology
Next.jsTypeScriptOpenAIPostgreSQLpgvectorNode.jsSupabaseVercel
