Skip to content
Back to Projects
OngoingAI/MLBackendFull-Stack

AI Collaborative Workspace

Full-stack graduation project with a context-aware AI assistant and multi-source RAG pipeline for organizational collaboration.

FastAPIRAGpgvectorReactPostgreSQL

The Problem

Organizations need AI assistants that understand their specific context, not generic chatbots. Knowledge is scattered across documents, databases, and team communications.

Architecture & Approach

Building a multi-source RAG pipeline with PostgreSQL/pgvector for grounded answers from organizational documents, files, and databases. FastAPI microservices architecture allows each service to scale independently.

Key Technical Decisions

pgvector over dedicated vector DB

Keeps everything in one database, simplifies deployment and operations. PostgreSQL is battle-tested for production workloads.

FastAPI microservices architecture

Each service (auth, documents, AI, collaboration) scales independently. A spike in AI queries doesn't affect document uploads.

Results

Ongoing graduation project with context-aware AI assistance for organizational collaboration.