Singapore – MongoDB has announced a set of updates to its Atlas platform at MongoDB.local Build Fest, aimed at improving how AI applications and agents retrieve information and connect to live operational data. The announcements cover new embedding and retrieval capabilities powered by Voyage AI, and a managed server that links MongoDB Atlas directly to AI coding tools.
New retrieval and embedding capabilities
The company introduced a new model in voyage-code-4, built for agentic code retrieval. The model, along with the full set of Voyage AI embedding and reranking models, is now available through a standalone API for use in any application. MongoDB’s Voyage AI embedding models are ranked on the Retrieval Embedding Benchmark (RTEB).
MongoDB also introduced Automated Embedding in Atlas, which is designed to remove the need for developers to manage separate embedding pipelines or keep vector stores in sync with operational data. Under the new setup, users set a Voyage AI embedding model for their search index, and Atlas embeds new documents as they are written and re-embeds existing ones when they change.
The company outlined four capabilities tied to the update: automated embeddings that keep context current as data changes; a single Atlas Embedding and Reranking API endpoint that any application can call; voyage-code-4 for code-specific retrieval; and vector search within Atlas Stream Processing for data still in motion.
Two customers were cited in connection with the update. The Financial Times consolidated search that had splintered across its teams and products onto the MongoDB platform, adopting Automated Embeddings powered by Voyage AI.
“Our job is to make the FT’s journalism fast and easy to reach, however our readers come to it. With Automated Embedding and Voyage AI models on Atlas, we’ve improved retrieval accuracy while keeping costs in check across more than 100,000 searches a day, and being able to test and balance models lets us tune quality against cost as we go. With less infrastructure to run, the team can spend more of its time on the reading experience our subscribers rely on,” said Elitsa Pavlova, Principal Engineer CM Platform, at Financial Times.
Eve, a legal AI platform, adopted the Atlas Embedding and Reranking API to surface relevant material as it works across the life of a case.
“In legal AI, retrieval quality is foundational—the right evidence has to surface at the right moment. MongoDB’s Atlas Embedding and Reranking API gives us a promising way to improve relevance directly in the RAG layer, while simplifying the infrastructure needed to build and evolve these experiences,” said Urvesh Patel, Staff AI Engineer at Eve.
“Too many organizations are running AI in production with an operational database, a vector store, a search engine, and embedding and reranking models, all from different vendors, bolted together instead of built for it,” said Jim Scharf, Chief Technology Officer, at MongoDB. “That’s where stale data and errors creep in, and it’s usually where teams spend their time babysitting instead of building. Agents raise the bar. They need to retrieve live context continuously and cannot wait on overnight batch jobs. MongoDB was built as an operational platform from the start, so retrieval and memory run on the same live data, nothing to sync, and agents act on what’s happening instantly.”
MongoDB Atlas Managed MCP Server for coding agents
Separately, MongoDB announced that its data platform is now available natively inside AI tools used by developers to build applications. The company launched the MongoDB Atlas Managed MCP Server, a fully hosted way to connect agents to Atlas without running additional infrastructure. Through the new connectors, developers can add MongoDB Atlas to Claude Code, Codex, Grok Build and Devin.
“The AI tools teams reach for keep changing, so our approach is to make sure MongoDB is present in all of them, whether a team is working in Claude or Codex, or running an agent in production. More of that building is now done by agents, and neither the agent nor the developer has to stop and set up a connection, so applications come together faster,” said Pablo Stern-Plaza, Chief Product Officer, AI and Emerging Products, MongoDB.
With the new connectors, developers can query data in plain language through ChatGPT, Claude and Grok; query, inspect and update data through coding agents such as Claude Code, Codex, Grok Build and Devin as work happens; and view live MongoDB data while generating an application in an IDE such as Cursor. MongoDB said connecting the tools requires only a few steps: finding the MongoDB connector in a tool’s marketplace and authorizing it, without pasting a connection string or configuring infrastructure. Once connected, a tool can list collections and indexes, query and aggregate data, and inspect schemas, and — with the right permissions — create collections or manage indexes.
MongoDB said its existing MCP server sees more than 30,000 installs a week. The new Managed MCP Server runs as a hosted service inside Atlas, removing the need for teams to install, operate or upgrade it themselves, and uses the same credentials and access controls teams already use with Atlas.
“Developers want their AI tools to connect with the context and systems they already rely on,” said Vibhor Chhabra, Product Lead for ChatGPT Ecosystem at OpenAI. “MongoDB’s plugin in ChatGPT makes it easier to access and work with live application data, helping developers move faster while staying grounded in the context of their applications.”
“We’re in the golden age of software engineering. The scope of what one engineer can build has exploded, and the unlock is agents working with real context,” said Russell Kaplan, President at Cognition, the company behind Devin. “By connecting Devin to MongoDB Atlas, engineers can hand off well-scoped tasks knowing Devin is working from live application data, not stale assumptions, and spend their own time on the harder problems.”
