AI agents are becoming increasingly active participants in data management, allowing users to interact with their data using natural language queries. A multi-agent AI system is proposed, where multiple agents work together to accomplish tasks, each with a specific role and communicating with others to share information and coordinate actions. The system combines four technologies: Azure OpenAI GPT-4o for understanding natural language, LangChain for converting text-based user requests into structured database queries, AutoGen for managing multi-agent orchestration and coordination, and Neon Serverless Postgres for storing CRM and shipment data. By using these technologies, users can build a simple yet powerful AI solution that not only retrieves data but also performs actions, such as creating new shipments or adding customers to the system. The proposed system allows for low-code development of agentic workflows and provides features like branching and row-level security to manage AI agents and their access to data.