MongoDB Integrates Google Cloud Vertex AI Machine Learning Platform to Atlas

Developers can now use Google Cloud’s Vertex AI to build generative AI applications running on the MongoDB Atlas cloud database. Vertex AI, Google Cloud’s managed machine learning platform, is now natively integrated with MongoDB Atlas.

MongoDB announced it has expanded its partnership with Google Cloud on Thursday at the MongoDB.local NYC conference in New York. The alliance builds on a partnership the two companies formed in 2018.

MongoDB’s Alan Chhabra

Core integration of Vertex AI with MongoDB Atlas is available. Still, it’s evolving, meaning the companies are building on the capabilities of Google Cloud’s large language models (LLMs), according to Alan Chhabra, MongoDB’s executive VP for worldwide partners and international sales.

“Eventually, it’s going to evolve and get smarter,” Chhabra told Channel Futures during an interview at the MongoDB event. “AI is still in the early days, and the capabilities from the cloud providers are evolving every day.”

Chhabra noted that integration will ease the ability to add of generative AI to MongoDB Atlas running in Google Cloud among MongoDB’s systems integrator and ISV partners.

Google Cloud is the first of the major cloud providers MongoDB has partnered with to integrate LLMs. Chhabra indicated that announcements of other alliances would be forthcoming. “This one was ready first, but others are in progress,” he said. “In the end, we’re cloud agnostic. But Google Cloud was very excited to work with us on this.”

Vertex AI Update

A Vertex AI update released earlier this month added generative AI modeling capabilities to the platform. The update gives developers access to Google’s various generative AI large learning models, including its text model PaLM 2, Embeddings API for text and foundation models in the company’s Model Garden. The models are also available in Google’s Generative AI Studio for model tuning and deployment.

“Backed by enterprise-grade data governance, security, and safety features, Vertex AI can make it easier than ever for customers to access foundation models, customize them with their own data, and quickly build generative AI applications,” wrote Google Cloud VP for cloud AI and industry solutions June Yang.

Developers seeking to add generative AI capabilities to their apps can start using the LLMs from Google Cloud’s tools to train models running in the MongoDB Atlas cloud-native, operational database. Chhabra said Google Cloud has worked closely with MongoDB on this integration. “Google wants us to be there too because Atlas drives Google Cloud consumption,” he said. “It’s a natural fit.”

Looking ahead at its partnership with Google Cloud, Chhabra envisions a future when developers will use AI to modernize applications by refactoring code, such as a COBOL DB2 application to Java, on a microservice architecture running MongoDB Atlas. Many systems integrators are looking to accelerate the ability to do that, Chhabra emphasized. “We want to work with them on that strategy,” he said.

The two companies are enabling the text embedding API from Google Cloud Vertex AI to gather data stored in MongoDB Atlas. It also uses Google Cloud’s PaLM text models to create capabilities such as semantic search, data classification, outlier detection, AI-capable chatbots and text summarization.