LAUNCHMongoDB 8.3 is built for the sub-100ms retrieval & zero downtime AI demands. Read blog >

AI DATAStop fighting your data layer. Get the memory & retrieval agents need to scale. Read blog >

DevRev

Enhancing AI data platform efficiency for DevRev with MongoDB

Industry

Computer Software

Product

MongoDB Atlas

Atlas Vector Search

Use Case

Gen AI

Single View

Customer since

2021

DevRev: Enhancing AI Data Platform Efficiency with MongoDB — Learn how Steven Poitras, Founding Architect at DevRev, built AgentOS on MongoDB Atlas and achieved low-latency performance for customers around the world.

THEIR CHALLENGE

Using MongoDB Atlas to consolidate data and boost efficiency for DevRev

The mechanics of running a modern business are complex, involving many separate teams and systems — from customer relationship management (CRM) and support to IT service management (ITSM). However, these commonly operate in silos, leading to inefficiency and communication gaps.

DevRev was born out of this need to bring data and people together, embracing the capabilities of AI and creating a new vision of a unified data platform. The company wanted to kick start innovation by centralizing data from previously isolated tools, including those for support chat, ticketing, CRM, engineering management, and ITSM.

“One of the main challenges was improving how we can coordinate across geographically dispersed teams and disparate systems,” said Steven Poitras, Founding Architect at DevRev. “By bringing data together and working on it in real time, we keep everyone in sync. Combine a platform built for AI with a full picture of the business and you have a gold mine for efficiency.”

Using DevRev’s AgentOS platform, all facets of a business can work together to drive productivity and help business leaders make objective, data-informed decisions. Intelligent AI models analyze and respond to a company’s complete context, so there are no more knowledge gaps or interruptions for updates.

DevRev’s platform is powered by MongoDB Atlas, a fully managed, multi-cloud, modern data platform. DevRev uses MongoDB Atlas to consolidate, aggregate, and converge all its platform, product, and user data to power AI agents — components that work in coordination to perform tasks. And the company aims to provide these AI agents across support, product engineering, operations, sales, and marketing. So, instead of having to go to various teams to ask about ROI and metrics, all teams can access the data in one place with AgentOS.

AgentOS handles billions of requests from customers per month for everything from AI-assisted insights and analytics to internal communications and development. Using MongoDB Atlas, the company built its platform to be scalable and flexible so that it would fit the needs of all its customers. “From the beginning, we always knew that MongoDB was going to be the core platform from a data perspective,” said Poitras. “When we looked at MongoDB, we saw that not only did it provide scalability and flexibility as a service, but there’s also a lot of innovation taking place.”

“With MongoDB Atlas, our development velocity is 3 to 4 times higher than if we used alternative databases. We can get our innovations to market faster, providing our customers with even more modern and useful CRM solutions.”

Anshu Avinash
Founding Engineer, DevRev

OUR SOLUTION

Improving development velocity by 3 to 4 times using MongoDB

Relying on MongoDB Atlas has empowered DevRev to innovate and make its AgentOS product the go-to solution that companies of all sizes can use to consolidate their teams and data sources into one platform.

“When you have tons of data and you have to move it from system to system, that’s very inefficient,” said Poitras. “We wanted one platform that would give us all the benefits in terms of flexibility, scalability, security, and performance and also had a lot of native capabilities built in.” DevRev not only minimized costs by adopting MongoDB Atlas, but it also optimized its efficiency and speed to market. “With MongoDB Atlas, our development velocity is three to four times higher than if we used alternative databases,” said Anshu Avinash, Founding Engineer at DevRev. “We can get our innovations to market faster, providing our customers with even more modern and useful CRM solutions.”

In addition, the company is powering AgentOS’s data aggregation and intelligent knowledge graph with MongoDB Atlas Vector Search. By storing vector embeddings in MongoDB Atlas and enabling semantic search with Atlas Vector Search, DevRev’s developers can augment the large language models that drive the AI agents and search features with relevant, domain-specific context from the semantic search results. This delivers more insightful and accurate responses, enhancing the user experience.

The company uses a microservices-based architecture to power AgentOS. To streamline how it sorts and queries large numbers of documents with similar structures, DevRev uses the MongoDB Attribute Pattern. With this pattern, it can quickly add, remove, or modify custom fields in these documents and more efficiently handle the wide variety of data structures that it uses to run AgentOS.

For DevRev, performance isn’t just about scaling resources to match customers’ needs. It’s also about scaling efficiencies, automations, and staff productivity. “AgentOS needs to scale both horizontally and vertically to meet customers’ needs,” said Poitras. “MongoDB offers both of those advantages.” And, because MongoDB Atlas supports a multi-cloud architecture, DevRev can benefit from expansive flexibility. The company isn’t locked in to any one choice of cloud provider, so it can build with the confidence that it will be able to change or add to its cloud architecture in the future.

As a smaller company, DevRev values the efficient resource use of MongoDB. It also enhanced agility by using the cloud-native features of MongoDB to keep up with the rapid rate of technological change. And by unlocking the power of automation, DevRev is being proactive about engineering for the future. “We can make people’s lives so much easier by offloading things to an AI system or using automations,” said Poitras.

“We’ve built a platform onto which we can plug and play additional agents as we grow and scale on MongoDB. That’s the thing with AI: you don’t really know what the future holds. We made something that’s flexible and foundational.”

Steven Poitras
Founding Architect, DevRev

THE OUTCOME

Building a scalable foundation for AI-powered efficiency

As DevRev has grown, it has evolved its use of MongoDB from simply using the service to communicating with the MongoDB team to develop new features and capabilities. The company chose MongoDB because it offered scalability along with ongoing innovation, so it will continue to support DevRev’s technical and business goals. “There’s a lot of collaboration to build not only MongoDB but also DevRev together and see how we can grow these solutions synergistically,” said Poitras. And DevRev continues to prioritize security, building multitenancy into AgentOS from the beginning and automating processes to gain consistency and make code review simpler.

On MongoDB, DevRev is providing a scalable, reliable platform for customers of all sizes and all locations to centralize their data and unlock better efficiency using AI. And the company has designed its foundation so that it can scale seamlessly as it grows.

To learn more, visit MongoDB Atlas.

Take the next step

Get access to all the tools and resources you need to start building something great when you register today.

Get StartedTalk to an expert