Hi, I'm Naresh Sampara (PhD).

2025-04 – 2025-08

Enterprise Multi-Agent RAG & System Recommendation Platform

This project presents an advanced Multi-Agent Retrieval-Augmented Generation (RAG) architecture built to unify and streamline diverse data retrieval needs across the CCS (Crown Commercial Service) ecosystem. Instead of relying on isolated search systems, this platform centralizes enterprise queries by intelligently routing requests to domain-specific agents. It seamlessly handles complex queries across internal HR and finance data, general CCS informational inquiries, and dynamic recommendation workflows for procurement Frameworks and G-Cloud services.

• Python
• LangGraph
• RAG
• AI agents
• Flask

Memory Types in AI Agents

AI agent memory architectures rely on six distinct systems to manage context, persistent knowledge, and execution skills. Short-Term (Working Memory) holds temporary dialogue context within the active LLM window, while Long-Term Memory persists user state across separate sessions via external databases. Semantic Memory tracks discrete facts and user preferences, whereas Episodic Memory logs chronological event history so the agent can evaluate past outcomes and avoid repeating mistakes. Finally, Procedural Memory defines operational rules and workflow skills for executing multi-step tasks, and Vector/Retrieval Memory leverages dense embeddings to fetch relevant unstructured knowledge on demand via semantic similarity search.

• AI
• AI agent
• Memory