MongoDB Launches Atlas Agent Engine to Accelerate Production AI Agent Deployment

MongoDB Tackles Production AI Agent Deployment Bottleneck

MongoDB unveiled Atlas Agent Engine at its Investor Day at the Nasdaq MarketSite in New York City, addressing a critical challenge facing enterprises: moving AI agents from proof-of-concept testing into production environments1

1

2

2

. The new platform provides a unified execution, memory, and governance layer specifically designed to overcome data retrieval, persistent memory, and security hurdles that typically prevent production AI agent deployment3

3

. Available today in public preview, Atlas Agent Engine eliminates the false tradeoff between adopting a single vendor's runtime with model and cloud lock-in, or piecing together disparate frameworks while managing governance independently5

5

.

Unified Platform Ends Vendor Lock-In Concerns

Atlas Agent Engine delivers retrieval powered by MongoDB Voyage AI, whose embedding and reranking models rank among top performers on RTEB, a benchmark reflecting real enterprise retrieval rather than academic datasets2

2

. The platform offers consumption-based pricing for Atlas Agent Runtime and Atlas Agent Memory, with usage drawing on customers' existing MongoDB Atlas commitments rather than requiring new contracts3

3

. Pablo Stern-Plaza, Chief Product Officer for AI and Emerging Products at MongoDB, emphasized that enterprises now get real-time context for their AI agents with governance and security built in from the start, plus freedom to run any model, any framework, and on any cloud2

2

. Customers can adopt the memory and governance layers independently or with the runtime, using existing models and frameworks they already know5

5

.

India R&D Hub Powers Core Development

MongoDB's rapidly expanding R&D hub in India played a central role in developing Atlas Agent Engine, with the engineering team growing to approximately 110 engineers since launching 18 months ago2

2

. Working alongside colleagues in the U.S., Europe, and Australia, the India team has evolved into a strategic global R&D engine, owning critical pieces of MongoDB's platform end-to-end3

3

. The team took on end-to-end product ownership of critical architectural components for Atlas Agent Engine and delivered key capabilities for MongoDB 9.0, including mission-critical backup features3

3

. MongoDB's overall regional footprint in India has grown to more than 750 employees across Gurugram, Bangalore, and Mumbai offices, serving 50 of the top 100 Indian companies by market capitalization and more than 50 local unicorn companies2

2

.

MongoDB 9.0 Delivers Major Performance Gains

Alongside Atlas Agent Engine, MongoDB released MongoDB 9.0, now generally available, offering substantial improvements in performance, scalability, availability, and security for modern AI-driven applications1

1

. The release provides up to 35 percent faster find-one queries compared to MongoDB 8.0, up to 30 percent faster update-one queries, up to 2x throughput on large instances, and up to 20 percent higher throughput for transactional workloads1

1

. MongoDB 9.0 includes enhanced industry-first queryable encryption capabilities targeted for applications processing highly sensitive data1

1

. The company also introduced Atlas Infinite, currently in public preview and described as the largest architectural innovation to MongoDB Atlas since its 2016 launch, which separates compute and storage functions for independent scaling1

1

.

Cognition Partnership Accelerates Legacy Infrastructure Migration

MongoDB partnered with Cognition to launch Devin for MongoDB Modernizations, automating legacy infrastructure migration to MongoDB Atlas4

4

. The integration combines Devin, an AI software engineering tool from Cognition, with MongoDB's Application Modernization Platform to automate code rewriting and data migration processes that typically require manual engineering work4

4

. In early testing, tasks previously requiring five to six hours were completed in just over an hour4

4

. MongoDB CEO Dev Ittycheria stated the integration allows enterprises to modernize larger portions of their application infrastructure simultaneously, moving decades of legacy applications onto MongoDB in months instead of years4

4

. Cognition has raised over $2 billion at a $48 billion valuation, and the service is available to MongoDB and Cognition customers starting today4

4

. MongoDB also announced it is joining the Linux Foundation's Open Secure AI Alliance and Agentic AI Foundation to support open software and standards for AI agents5

5

.