Powering Event-Driven, Multi-Agent AI: Confluent Named MongoDB Global Tech Partner of the Year

We’re proud to announce that Confluent has been named MongoDB’s 2025 Global Tech Partner of the Year. This award highlights the strength of our partnership and joint go-to-market execution, helping enterprises build the next generation of intelligent, event-driven artificial intelligence (AI) applications.

MongoDB and Confluent deliver a powerful AI solution by combining Confluent’s complete data streaming platform, unifying Apache Kafka® and Apache Flink®, with MongoDB’s operational data storage, advanced vector search, and embedding capabilities. Together, they enable enterprises to seamlessly source, process, enrich, govern, and serve data from every corner of the business to power both agentic and generative AI applications.

Deploying Event-Driven, Multi-Agent AI Systems

Without the right architecture, building multi-agent systems can become the next-generation monolith: brittle, expensive, and hard to evolve. The future of agentic AI is event-driven—decoupled systems composed of independent, specialized agents communicating asynchronously and working collaboratively to solve complex problems more efficiently.

Together, Confluent and MongoDB provide the data foundation necessary to successfully bring AI workloads to production—quickly, reliably, and at scale. Streaming Agents on Confluent Cloud let you build, deploy, and orchestrate event-driven agents natively on Flink. MongoDB provides a rich, flexible data store with powerful vector search and a global, developer-friendly operational platform that ensures that agents always have access to the freshest, most relevant context.

The result: a robust, decoupled, and continuously updated multi-agent ecosystem that can perceive, reason, and act on live business data in real time.

Providing Fresh, Accurate Context for RAG

Streaming Agents can transform unstructured enterprise data into real-time vector embeddings. Combined with MongoDB Atlas Vector Search, these embeddings allow your AI agents to instantly retrieve the most relevant information to improve decision-making, enable retrieval-augmented generation (RAG), and mitigate hallucinations. A variety of embedding models—OpenAI, Amazon Bedrock, Google Gemini, MongoDB Voyage AI, and more—can be used across any cloud, while the Create Embeddings Action provides a no-code shortcut to keep context continuously fresh.

The fully managed MongoDB Atlas Sink Connector ensures that enriched and vectorized data flows seamlessly between Confluent Cloud and MongoDB Atlas, keeping all systems in sync without manual maintenance.

Enhancing Data Enrichment for More Accurate AI Decision-Making

Multi-agent systems need more than raw data. They require context from across the enterprise in order to act on the most current, complete view of what’s happening in the business. Flink SQL on Confluent can join real-time data streams with operational data in MongoDB Atlas as well as relational databases and REST APIs. This allows agents to leverage both vector search for RAG and fast external table lookups without complex synchronization pipelines.

Learn more about Streaming Agents on Confluent Cloud.

Real-World Impact: A Multi-Agent Complaint Resolution System

For retail banks overwhelmed by high volumes of routine customer complaints—such as card declines, authentication issues, or billing questions—a multi-agent AI system built with MongoDB and Confluent automates classification, contextual reasoning, and resolution in seconds, reducing manual effort and improving customer satisfaction. In this solution, Confluent streams trigger and update the flow of customer inquiries in real time while MongoDB Atlas provides the operational backbone and vector search capabilities that allow specialized AI agents to classify, prioritize, and resolve cases faster than ever.

The result is a smarter, more automated customer experience that reduces operational costs and accelerates resolution times.

Check out the full article on MongoDB’s blog.

Confluent: The World’s Data Streaming Platform

Confluent was founded by the original co-creators of Apache Kafka to make data in motion accessible to every organization. Today, it offers the most complete, cloud-native data streaming platform to stream, connect, process, and govern data across any cloud or environment. When pairing it with MongoDB’s versatile, developer-friendly database platform, enterprises can gain a foundation for building modern, data-driven applications and AI systems that scale globally, reliably, and securely.

By eliminating the operational burden of self-managing Kafka, Confluent Cloud can reduce total cost of ownership (TCO) by up to 70% while accelerating time to market for new AI applications.

Start your free trial of Confluent Cloud. New sign-ups receive $400 to spend during their first 30 days.

See the Impact Live at Current 2025

Want to learn more? Join us at Current 2025, where real-time data and AI innovation converge. Discover how Confluent and MongoDB together are reshaping how organizations build intelligent, event-driven, multi-agent systems for the future.

Register Now

Apache®, Apache Kafka®, Kafka®, Apache Flink®, Flink®, and the Kafka and Flink logos are registered trademarks of the Apache Software Foundation. No endorsement by the Apache Software Foundation is implied by the use of these marks.