10 Years of MongoDB Atlas: Built for What’s Next

Nearly a decade ago, I joined MongoDB as a Senior Product Manager to help build the company’s new cloud product, MongoDB Atlas. Our customers had been telling us they wanted to bring MongoDB’s familiar developer experience to the cloud, with the reliability and confidence teams needed to run in production. Atlas was our answer.

Today, we’re celebrating 10 years of MongoDB Atlas, the generational data platform for AI applications, and the customers who pushed us to build it.

Atlas was shaped in close conversation with those customers and scaled alongside them every step of the way. Today, more than 250,000 builders get started on Atlas every month. Atlas serves more than three trillion queries a day (a roughly threefold increase just since 2023!), and represents 75% of MongoDB’s revenue. Those numbers reflect something more important than growth: the trust builders and customers have placed in us to scale their businesses.

That trust was earned by listening closely. Every major capability and architectural investment in Atlas was rooted in what customers asked for: the flexibility and speed of MongoDB’s document model, delivered in a platform that removed operational overhead and could scale with their applications. Over time, Atlas expanded beyond a managed database into a broader data platform, because builders kept asking for more flexibility, more simplicity, and more room to build.

That matters even more in the AI era. AI applications create new demands, but the underlying requirement is familiar: builders need a platform that can support operational data, search, and retrieval while scaling through constant change—without forcing them to stitch together a mess of disconnected systems. We spent ten years becoming the flexible, durable data platform that builders trust. Those are the same qualities AI applications need most, and that’s why builders are now using Atlas to build trustworthy AI applications with highly accurate retrieval, real-time context, and the scale to run in production.

Managed cloud databases become the default

When Atlas launched in 2016, organizations were moving away from traditional data center build-outs and toward cloud-based delivery, a market Gartner forecasted would reach $204 billion (and is now approaching $1 trillion).

Developers loved MongoDB as a flexible, intuitive foundation for building applications, but they also wanted to take advantage of the cloud. Atlas’s first promise was simple: bring MongoDB’s familiar developer experience to the cloud, with the reliability and confidence teams needed to run in production.

To deliver that confidence to developer teams, we built Atlas with security, resilience, and performance at its core—from encryption and access controls to backups and high availability. The result was a service that teams could run in production with confidence, freeing developers to do their very best work without the headaches associated with database administration.

By 2018, 81% of enterprises were operating in multi-cloud environments, and an IDG study found that more than half indicated they were thinking about cloud as a portfolio strategy. As customer architectures became more distributed, teams needed the flexibility to choose the cloud environment that fit their applications, teams, and compliance needs.

To support them, we extended our original promise of simplicity into multi-cloud flexibility, with availability across all three major cloud providers. And in 2020, we introduced Atlas Multi-Cloud Clusters, making Atlas the first and only cloud database to let customers run applications simultaneously across AWS, Azure, and Google Cloud regions—a unique achievement that gave organizations that require ultra-high availability one consistent data foundation across all the major clouds.

Today, customers can run across over 125 AWS, Google Cloud, and Microsoft Azure cloud regions, making Atlas the most widely available managed data platform in the world.

Enterprises' scale, and consolidation becomes a customer priority

As cloud adoption accelerated, customers wanted more than a hosted database. The cloud had become a long-term investment, and developers needed global reach, resilience, and a platform that could handle more workloads, securely, without requiring them to keep adding infrastructure around it. Because developers already trusted us on the fundamentals, Atlas could expand into the kinds of workloads enterprises could not afford to get wrong.

For workloads like payments, inventory, and order processing, strong transactional consistency is a requirement. The addition of multi-document ACID transactions in 2018 brought that transactional consistency to MongoDB and marked an important step in MongoDB’s evolution, enabling MongoDB to serve the kinds of high-stakes transactional workloads that enterprises had historically reserved for relational databases. Now, customers could use MongoDB with greater confidence for a wider set of systems where accuracy, resilience, and trust could not be compromised.

MongoDB extended its trustworthy database foundation with the launch of MongoDB Queryable Encryption, an industry-first encryption capability that allows customers to query encrypted data while keeping sensitive information protected when it is at rest, in transit, and in use—an important step for securing regulated and highly sensitive workloads.

At the same time, Atlas continued to evolve to help customers operate at a larger scale. In 2020, we introduced Atlas Search and Online Archive, adding rich application search and giving customers a simpler, lower-cost way to store older data without losing easy access to it. In 2021, Native Time Series Collections and Live Resharding followed, helping customers manage time-stamped data more efficiently and scale architectures without downtime.

These updates made Atlas easier for builders to work with as deployments became bigger, more distributed, and more complex, all while minimizing the number of disparate systems that development teams had to stitch together and maintain.

Trustworthy AI becomes the new frontier

Then, the launch of ChatGPT in late 2022—and with it the rise of generative AI—created a massive new challenge for builders.

Enterprise adoption moved faster than standards and controls, leaving teams to figure out how to connect the necessary data components to run semantic search and retrieval-augmented generation (RAG) workloads together without creating a brittle mess of data pipelines, sync jobs, and specialized infrastructure that compromised security and performance.

To help teams bring these critical AI building blocks together on one secure platform, Atlas evolved again. With the public release of Atlas Vector Search in 2023, MongoDB was one of the first databases to launch vector search as a native capability, which enabled developers to keep vectors close to operational data and run semantic retrieval directly in the database without having to manage a separate vector store. Search Nodes gave teams a way to scale search and vector workloads independently from the operational database, while Atlas Stream Processing gave builders a way to process real-time streaming data without adding separate infrastructure.

The business demand for this architecture has been staggering: over 726,000 vector indexes and 55,000 vector applications have been created since we introduced Atlas Vector Search, and we’ve seen a 92% increase in customers showing production-level vector search usage over the past 12 months.

And with the company’s acquisition of Voyage AI in 2025, MongoDB sharpened its focus on retrieval quality—bringing advanced embedding and reranking models into Atlas. The integration of Voyage AI was about rethinking the data architecture to help customers reduce hallucinations, improve relevance, and make AI useful in the real-world environments where accuracy and trust matter most.

This immediately paid huge dividends for customers building highly accurate semantic search and RAG applications. But we knew that as the market moved towards autonomous AI, trustworthy retrieval and access to real-time context would matter even more.

Agents and the future of the data layer

Today, we’re firmly in AI’s agentic era. Builders want to deploy agents that can reason over business context with autonomy. But agent memory requires fast accuracy at scale so that the right information is recalled at precisely the right time. And this is where they run into a challenge. They're excited about agents, but they can't put an agent in front of their customers if the results are inconsistent, irrelevant, or flat-out wrong.

That puts increasing focus on the data layer of the tech stack. Agents are only as good as the context they can retrieve, rank, and retain. If the underlying data is stale, incomplete, or poorly retrieved, the output will be wrong—regardless of how strong the model is. In practice, production agents depend less on model choice alone than on retrieval quality and the ability to ground responses in live operational data.

With search, vector search, embeddings, and rerankers natively integrated into the Atlas platform, businesses are closing the gap between data and retrieval to produce fast, accurate results for agents at scale. And with foundational capabilities to ensure exceptional security, resilience, and performance, builders are freed up to do what they do best, instead of spending their days bogged down managing data infrastructure.

Over the past decade, our goal has been to reduce operational burden for customers without compromising on the technical bar. As the industry moves toward agents, that aim still applies.

We’re ten years in, and Atlas has grown into the data platform that runs intelligent, mission-critical applications for nearly 70,000 customers across every industry. The world runs on Atlas! Our customers pushed us to build everything that matters in the platform, so they could do more, faster. The same holds true today: the agentic AI era is raising the bar for innovation, and we're raising it with them. The ambition our customers bring to what they're building next is what drives us forward—and we're ready for it.

Here's to the next 10 years.