Full Agent capabilities with MongoDBJoin us for an evening dedicated to **practical AI applications**, where we’ll explore how to build intelligent agents and harness the power of vector search. Whether you’re a developer, data enthusiast, or simply curious about AI, this meetup is a great opportunity to learn, connect, and have fun with the community.
**Agenda**
**6:30 PM – Welcome, Registration & Meet and Greet**
Arrive early, grab your badge, and connect with other AI enthusiasts before the sessions begin.
**7:00 PM – Running a RAG locally**
Learn how to build and run Retrieval-Augmented Generation (RAG) systems entirely on your local environment, ensuring privacy, control, and reduced latency.
**7:30 PM – MongoDB MCP**
Discover how MongoDB integrates with the Model Context Protocol (MCP) to enable scalable, real-time data access for AI applications and agents.
**8:00 PM – Long-term memory for agents**
Explore strategies and architectures to equip AI agents with long-term memory, enabling more consistent, context-aware, and personalized interactions.
**8:30 PM – Pizza, Networking, Swag & Closing**
Enjoy some pizza, grab exclusive MongoDB swag, and continue networking with the community.
**Sessions**
**Title: Running a RAG locally**
**Summary:** In this session, we will walk through how to design and deploy a Retrieval-Augmented Generation (RAG) system running fully on local infrastructure. We’ll cover key components such as embeddings, vector databases, and local LLMs, and discuss trade-offs between performance, cost, and privacy. Attendees will gain a practical understanding of how to build secure and efficient AI systems without relying on external APIs.
**Title: MongoDB MCP**
**Summary:** This talk introduces MongoDB’s role within the Model Context Protocol (MCP) ecosystem, enabling AI models and agents to interact seamlessly with live data. We’ll explore how MCP standardizes communication between models and tools, and how MongoDB can be used to store, retrieve, and manage structured and unstructured data in real time. Practical use cases and integration patterns will be demonstrated.
**Title: Long-term memory for agents**
**Summary:** As AI agents become more sophisticated, the need for persistent and meaningful memory becomes critical. In this session, we will explore different approaches to implementing long-term memory, including vector-based memory, structured storage, and hybrid architectures. We’ll also discuss challenges such as memory retrieval, relevance, and scalability, and how these impact real-world agent performance.
**Important:** This event will be held in Spanish / Este evento se realizará en español