It’s practically the tip of the 12 months—once more! Meaning it’s time for an end-of-year weblog publish that expresses disbelief on the passage of time. Which, because the saying goes, flies whenever you’re having enjoyable. Time additionally zips by whenever you’re busy, which MongoDB definitely was in 2025.
It was an enormous 12 months for the corporate—and extra importantly, for the tens of 1000’s of consumers and thousands and thousands of builders who depend on MongoDB’s trendy information platform for his or her most mission-critical workloads. At MongoDB, every thing we do begins with our obsession with prospects and their wants, and if there’s a theme to MongoDB’s 2025, it was (and can proceed to be) enabling buyer innovation and serving to them succeed within the AI period.
So listed here are a number of highlights of how MongoDB constructed for (& on behalf of) prospects in 2025. From the acquisition of Voyage AI to buyer success throughout industries, quite a bit occurred in 2025. Let’s go!*
*Learn to the tip for 2026 ideas.
2025: The (MongoDB) 12 months that was
Voyage AI, modernization, and search
In February, MongoDB introduced the acquisition of Voyage AI, a pioneer in embedding and reranking fashions, to boost the accuracy of AI purposes. Integrating Voyage AI’s superior retrieval know-how with MongoDB’s trendy, AI-ready information platform addresses a crucial problem: LLM mannequin hallucinations brought on by a scarcity of context. By bettering retrieval accuracy for specialised domains like finance and legislation, the combination allows companies to deploy AI for mission-critical use circumstances.
To be taught extra, see the MongoDB Voyage AI web page.
Then, in September, we launched MongoDB AMP, an AI-powered Software Modernization Platform. AMP is designed to speed up the transformation of legacy purposes by a mixture of AI-powered tooling, a confirmed supply framework, and knowledgeable steerage (instruments, strategies, and expertise) to assist enterprises cut back technical debt and modernize 2-3 occasions sooner.
Need extra? Positive you do! Take a look at this quick video.
MongoDB additionally introduced the addition of search and vector search capabilities to MongoDB Neighborhood Version and MongoDB Enterprise Server. This permits builders to construct and take a look at AI-native purposes, together with these utilizing retrieval-augmented technology (RAG), in native or on-premises environments. Beforehand unique to MongoDB Atlas, these options allow safe, hybrid deployments the place delicate information can stay on-premises whereas nonetheless leveraging superior search instruments.
Right here’s a (barely much less quick) video about search and vector search on Enterprise Server.
Rising and scaling with MongoDB
As famous, every thing we do at MongoDB begins with our obsession with prospects. 2025 was one other banner 12 months for buyer success and innovation—we had been impressed by what organizations of each form and dimension, throughout industries and geographies, constructed with MongoDB in 2025. Listed below are simply two of the numerous tales our prospects shared in 2025; far more could be present in my colleague Katie Palmer’s weblog collection, Innovating with MongoDB.
Manufacturing facility
By combining the Atlas trendy information platform with Voyage AI’s high-performance embeddings, the AI-native startup Manufacturing facility—which makes use of AI brokers known as Droids to speed up software program improvement lifecycles for organizations—consolidated its fragmented tech stack. This enabled superior code retrieval, simplified operations, and supplied the scalability wanted to course of billions of tokens day by day.
McKesson
McKesson, a world pharmaceutical distributor, changed its monolithic legacy infrastructure with MongoDB Atlas to satisfy strict drug tracing mandates. By adopting our trendy cloud information platform, McKesson scaled its operations 300x, managing monitoring information for 1.2 billion containers yearly with out latency, and making certain compliance and affected person security whereas lowering developer complexity.
For extra, try the video of McKesson at MongoDB.native NYC from September.
From area of interest NoSQL to enterprise powerhouse
As senior MongoDB engineer and Technical Fellow Ashish Kumar put it earlier this 12 months, “by a sustained and deliberate engineering effort,” MongoDB has gone from a (seemingly) area of interest NoSQL resolution to a trusted enterprise commonplace, and now delivers “the excessive availability, tunable consistency, ACID transactions, and strong safety that enterprises demand.”
A brand new period of management
The face of MongoDB has additionally modified—our CFO, Mike Berry, joined the corporate in April, and Dev Ittycheria stepped down as CEO in November, after greater than 11 years main the corporate (together with its 2017 IPO). In a LinkedIn publish about his position, new MongoDB CEO CJ Desai famous that the corporate is “on the forefront of a brand new information revolution, unlocking the following wave of productiveness and intelligence.”
“Having spent my profession constructing and scaling know-how platforms, I’ve all the time been drawn to corporations outlined by readability of imaginative and prescient, relentless natural innovation, and a customer-first tradition. MongoDB exemplifies all three,” stated Desai.
We couldn’t agree extra. Onward!
Studying the 2026 tea leaves
So what would possibly 2026 deliver (for MongoDB and tech at giant)? Listed below are a handful of our leaders’ predictions:
“As a lot as individuals wish to speak about Synthetic Basic Intelligence (AGI), we’re nonetheless within the section the place most AI use circumstances automate redundant duties however profit from human-in-the-loop checks. Organizations that use AI to finish work that traditionally is a drain on human sources—however then makes use of individuals to rigorously confirm what AI builds, apply governance frameworks, and preserve accountability throughout the information lifecycle—will probably be extra profitable.”
—Pete Johnson, Discipline CTO, AI, MongoDB
“After years of inflated expectations and unsustainable spending, the AI trade is trapped in a bubble the place corporations reflexively try to deploy LLMs at each drawback, driving up prices with minimal to no return. Companies that break away from this spending cycle are those that perceive the necessity to floor LLM responses in factual information and be taught from prior errors. We consider one of the simplest ways to do that will probably be with extremely correct embedding fashions and rerankers for dependable information retrieval.”
—Frank Liu, Employees Product Supervisor, MongoDB
“In 2026, cloud independence will evolve from strategic desire to existential crucial throughout enterprises of each scale. The outages and disruptions of current years have uncovered a basic fact: in an always-on digital economic system—the place commerce, mobility, governance, and even public security depend upon uninterrupted entry to cloud providers—single-provider reliance is not a calculated danger, however a systemic vulnerability.
Compounding that is the inexorable rise of knowledge sovereignty. Regulatory regimes worldwide now demand exact jurisdictional management over information residency, rendering inflexible cloud commitments incompatible with compliance at international scale.
The defining aggressive benefit will belong to organizations that transcend fragile prevention theater and engineer true infrastructural resilience: architectures inherently transportable, information frictionlessly cellular, and operations autonomously sustained throughout heterogeneous clouds by AI-orchestrated redundancy.
In brief, the winners won’t merely mitigate downtime—they’ll design programs that render the idea out of date.”
—Ben Cefalo, SVP, Head of Core Merchandise, MongoDB
Completely happy holidays and completely satisfied New Yr, everybody!
Subsequent Steps
Head over to our updates web page to be taught extra in regards to the new releases and updates from MongoDB in 2025.
Regulate our occasions web page to be taught what’s to come back from MongoDB in 2026, together with MongoDB .native San Francisco in January. Be part of us!
