The potential impact of Large Language Models (LLMs) on industries like manufacturing is significant, with estimated annual growth rates between 40% and 55%. LLMs can be applied across the value chain to optimize manufacturing operations, from inventory categorization to predictive maintenance strategy generation. However, truly transformative AI-powered applications need to evolve beyond chatbots and respond to user queries, acting on behalf of the user in complex processes. Agentic systems, which consist of multiple AI agents collaborating with each other, are emerging as the next frontier of generative AI applications, enabling objective-driven actions and context understanding. An agentic system can be customized to perform specific tasks, interact with humans for feedback, and optimize various facets of manufacturing operations simultaneously. MongoDB can act as a memory provider for such systems, leveraging its flexible document model, security features, and horizontal scalability. A proposed use case in manufacturing demonstrates the potential of multi-agent collaboration in optimizing machine performance by integrating predictive maintenance, process optimization, and quality assurance agents.