MongoDB launches Atlas Agent Engine for production AI

MongoDB has launched Atlas Agent Engine, a product for deploying AI agents in production. Available in public preview, it is aimed at customers already using MongoDB Atlas.

The software combines execution, memory and governance in a single layer to address a common challenge: moving AI agents from pilot projects into day-to-day use. It is designed to work with existing models and frameworks rather than requiring a separate technology stack.

Atlas Agent Engine launches alongside MongoDB 9.0 and Atlas Infinite, which MongoDB presented as related parts of its broader data and AI offering. The company said Atlas Agent Engine sits on top of those services and uses Voyage AI models for embeddings and retrieval.

The product uses consumption-based pricing for Atlas Agent Runtime and Atlas Agent Memory. Usage can be applied against existing Atlas commitments, allowing customers to adopt the software through infrastructure they already have in place.

Customer interest

MongoDB cited several early users and evaluators, including Australian mortgage and fintech group Lendi Group and payments company Paysafe.

"Designing marketing collateral for our retail locations is a manual process today, and it's one we think is well suited for agent-assisted support. We're evaluating MongoDB Atlas Agent Engine's ability to handle the complex, long-running, non-deterministic tasks like managing image layers and in-image compliance that our current agent platform struggles with. We see real potential to close that gap while bringing far more efficiency to how we produce marketing materials across our retail network," said Devesh Maheshwari, CTO, Lendi Group.

Paysafe highlighted a different use case focused on payments monitoring and analyst workflows.

"Investigating unusual activity in our payment network today means our analysts stitch together data from multiple systems by hand, often under time pressure. We're excited about the potential for an intelligent agent, built on MongoDB's Atlas Agent Engine, to shrink the time between a problem emerging and our team acting on it, giving our analysts more time to focus on the judgment calls that matter most," said Amar Akshat, SVP of Architecture, Paysafe.

Governance focus

MongoDB is positioning governance as a central feature of the launch. The platform logs actions against identities, whether human or software-based, and applies policy controls through a single control plane.

That approach reflects a broader market concern over how businesses supervise autonomous systems that can access company data, trigger workflows or make recommendations. Companies experimenting with AI agents have often built systems from separate tools for retrieval, memory, identity and monitoring, adding complexity when models or frameworks change.

James Governor, Co-Founder of RedMonk, said context and information management remain major issues for organisations trying to move into autonomous agent systems.

"Context is the critical success factor in successfully using agents for application development. Enterprises are currently struggling to assess, integrate and manage information across multiple systems to enable an ontology for autonomous agentic work," said James Governor, Co-Founder of RedMonk. "MongoDB Atlas Agent Engine is designed to bake governance into agentic app development with a single platform for memory and identity."

Open approach

Atlas Agent Engine is designed to avoid dependence on a single model, framework or cloud provider. MongoDB said it supports open standards including MCP and A2A, and can run across different cloud environments as well as self-managed infrastructure.

Pablo Stern-Plaza, Chief Product Officer, AI and Emerging Products at MongoDB, said the company built the product to address what it sees as a false choice between tightly integrated vendor platforms and more fragmented in-house approaches.

"Organisations that want to put agents in production are being forced into a false tradeoff: either adopt one vendor's runtime and accept being locked into a model and cloud, or piece together a framework and manage governance and memory on their own," said Pablo Stern-Plaza, Chief Product Officer, AI and Emerging Products at MongoDB. "With the launch of Atlas Agent Engine, that false tradeoff ends today. Enterprises get the real-time context their agents need, with governance and security built in from the start, and the freedom to run any model, any framework, and on any cloud. We didn't want to ask customers to predict the future. We wanted to build something that works no matter what they choose."

The launch also highlights MongoDB's effort to deepen its role in AI infrastructure beyond its core database business. The company said more than 70,000 customers already run on its platform, and it is tying Atlas Agent Engine to the broader Atlas environment rather than offering it as a standalone system.

Accenture was among the partners named in the ecosystem around the product.

"Atlas Agent Engine brings the enterprise-ready capabilities, context, and constraints needed to help AI agents deliver real-world impact. Combined with Accenture's governance, architecture, and deep industry expertise, it creates a powerful foundation for accelerating AI transformation and delivering outcomes at scale. Our shared commitment to delivery and customer success makes this partnership particularly strong," said Ram Ramalingam, Global Lead, SW Engineering & Head of RDE, Accenture.