The enthusiasm around generative artificial intelligence is escalating daily, with increasing confidence in the improvement of language models across the industry. Within this dynamic landscape, MongoDB Inc. has established itself as an essential hub for developers, frequently organizing hackathons.

About two years ago, the company launched MongoDB Ventures. The thought at the time was to look across the customer base and think about a good way to introduce new technologies to it, according to Suraj Patel (pictured), head of MongoDB Ventures.

“Generally, we’re looking at a lot of infrastructure software, a lot of developer tools,” he said. “The last year and a half has been very exciting … focused a lot on looking at companies building in the AI space and, in particular, companies that are enabling developers, which is our key audience, to build with AI.”

Patel spoke with theCUBE Research analysts John Furrier, executive analyst, and Dave Vellante, co-founder and chief analyst, at the “Supercloud 6: AI Innovators” event, during an exclusive broadcast on SiliconANGLE Media’s livestreaming studio theCUBE. They discussed the evolving landscape of developer-focused ventures in the AI space and the integration of AI into enterprise applications.

New AI solutions rooted in developer innovation

Something that has emerged as a subject of conversation during Supercloud 6 coverage on theCUBE involves a new formula that enterprises are driving toward. It’s end-to-end new kinds of systems to power AI but needing to be developer-led. It begs the question: How does one become an enterprise motion end-to-end system that is developer-led? Such trends are exactly why MongoDB is positioned well as a venture organization, according to Patel.

“What we’re seeing is that, really, the products that a developer needs to … build AI into their applications, even if they’re at the enterprise, those products are often startups,” he said.

A lot of developers within the MongoDB ecosystem are using products such as LangChain and LlamaIndex, among others, according to Patel. These are integral products for those companies that need to build AI-infused applications.

“Now, the data in an enterprise, though, lives in proven database technologies like MongoDB, and that’s where the magic of the partnership with some of these younger companies really comes into play,” he said. “We work hand in hand with them; we figure out what are the points of connectivity that we can enable and how do we make it easier for developers to use these new innovative companies with their existing enterprise data and really bring the right context to these AI applications.”

Anyone spending a lot of time in the AI space and talking to enterprise customers in it right now often hears the same narrative, according to Patel. Where it felt like last year was a lot of prototypes with people trying applications for the first time, things are now maturing.

“They’re starting to think about, OK, well, if I’m using this model as part of my application, what is the long-term cost going to be of this model? And how am I going to deal with that from a budget perspective?” he said. “If I’m using this framework, am I confident that this framework is now going to scale? Or if I’m introducing AI powers into my application, which are by nature probabilistic, how can I design unit tests or different testing to make sure that the actual output of the application is as predictable as possible, as well as safe?”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE Research’s coverage of the “Supercloud 6: AI Innovators” event:

Photo: SiliconANGLE