MongoDB Blog

Artificial Intelligence
Launched in 2017, Observe.AI was originally known for its expertise in conversation intelligence and contact center analytics. In recent years, however, the company has grown into a comprehensive agentic AI platform, and now has three core offerings:
The federated Copilot connector for MongoDB Atlas gives organizations a governed way to bring live operational data into the flow of work across Copilot experiences.
As applications grow, database teams need to make increasingly complex decisions about how to distribute data, manage workloads, and scale their infrastructure. Sharding enables MongoDB to scale horizontally, but managing a sharded cluster can introduce its own challenges. Understanding cluster health, diagnosing uneven load, and choosing an appropriate shard key can require piecing together metrics across multiple views and relying on specialized expertise.
Teams increasingly want agents that handle routine database work: checking cluster alerts, reviewing slow queries, and answering questions about a schema. But building agents takes much more than a prompt. It requires credential handling, access scoping, tool orchestration, approval flows, a chat integration, evaluation, observability, and ongoing testing and maintenance. That plumbing is easy to underestimate, and it's where much of the effort goes.
The MongoDB Extension for Hibernate ORM is now generally available, giving Java teams a production-ready way to use familiar Hibernate APIs with MongoDB’s document model. It is a first-party, open-source library that enables Java applications to use MongoDB as the backing database via the Hibernate ORM. The extension allows Java developers to enjoy the same experience with MongoDB, reducing the friction for those who want to switch.
Modern applications need to stay available even when traffic spikes suddenly or queries consume disproportionate resources. MongoDB Atlas is designed for high availability, but availability in practice is about more than surviving host failures or a replica set election. It’s also about how a cluster behaves when demand exceeds what it can safely process.
Modern applications generate telemetry across every layer of the stack, but understanding what changed and why often remains difficult. Database signals frequently live in separate workflows, forcing teams to switch between dashboards, correlate disconnected data, and piece together performance issues without the full operational context.
Last week, we announced the general availability of prefix, suffix, and substring query support in MongoDB Queryable Encryption. These capabilities extend Queryable Encryption beyond equality and range queries, and they’re now fully supported for production workloads.
For years, organizations have been told to accept a series of tradeoffs when it comes to their technology stacks:
Building AI applications has never been easier. Making them reliable in production remains the real challenge.
Every team building AI agents discovers one fundamental truth: without the right information, an agent cannot reliably answer a question or take the correct action.
MongoDB.local NYC 2026 is officially a wrap, and what an event it was!