New Delhi, May 7: MongoDB has announced a series of new capabilities at MongoDB .local London 2026, aimed at helping enterprises run AI agents in production through its unified AI data platform.
The announcements introduce native AI-focused capabilities including automated embeddings generation, persistent agent memory, enhanced database performance through MongoDB 8.3, and secure cross-region connectivity support for AWS PrivateLink. Together, these additions are designed to simplify AI infrastructure, improve retrieval accuracy, and accelerate the deployment of AI-powered applications at enterprise scale.
MongoDB said the latest innovations strengthen its strategy of offering enterprises a unified platform that combines operational data, vector search, full-text search, embeddings, memory, and reranker models within a single ecosystem.
“The hardest part of running agents in production isn’t the model. It’s the data layer underneath it,” said CJ Desai, President and Chief Executive Officer of MongoDB. “To trust an agent at scale, it has to retrieve the right context, hold memory across sessions, and operate at machine speed, wherever the enterprise needs it.”
Among the key announcements, MongoDB introduced Automated Voyage AI Embeddings in MongoDB Vector Search, now available in public preview. The feature automatically generates embeddings as data is written or updated, enabling AI agents to retrieve more accurate and real-time contextual information without requiring enterprises to build complex infrastructure manually.
The company also announced the general availability of LangGraph.js Long-Term Memory Store Integration, providing JavaScript and TypeScript developers with persistent cross-conversation memory capabilities powered by MongoDB Atlas.
Highlighting the importance of the data layer in AI systems, Pablo Stern, Chief Product Officer, AI and Emerging Products at MongoDB, said, “Developers no longer have to build and maintain data infrastructure, wire up embeddings, or manage syncing between systems. They can focus on business outcomes rather than the plumbing.”
MongoDB additionally unveiled MongoDB 8.3, now generally available, which delivers substantial performance gains over MongoDB 8.0, including higher read and write throughput, improved ACID transaction handling, and faster execution of complex operations without requiring application code changes.
The company also announced general availability of cross-region connectivity support for AWS PrivateLink, enabling secure database traffic between MongoDB Atlas clusters across AWS regions without exposure to the public internet. The feature is expected to support enterprises operating under strict compliance and data residency requirements.
Commenting on the product enhancements, Ben Cefalo, Chief Product Officer, Core Products at MongoDB, said, “MongoDB 8.3 makes agent workloads faster and cheaper to run on infrastructure customers already have. Production AI doesn’t wait, and neither do we.”
Additional announcements at MongoDB .local London 2026 included the general availability of Feast Feature Store Integration with MongoDB, new query expressions for data transformation, and MongoDB AI Skill Badges.
MongoDB said the latest developments reinforce its focus on enabling enterprises to build, deploy, and scale AI applications and intelligent agents through a single integrated platform.
