Skip to content

Search the site


MongoDB CEO: “It wasn’t AI...”

"I will clearly say it wasn't AI that drove the acquisition of workloads. It was really sharp execution by go-to-market teams.”

MongoDB AI opportunities

"People tend to overestimate the impact of a trend like AI in the short term, so I will clearly say it wasn't AI that drove the acquisition of workloads. It was really sharp execution by go-to-market teams.”  

Despite these forthright words from MongoDB CEO Dev Ittycheria after reporting strong earnings this week (words that will no doubt have been noted and appreciated by his employees), the markets headlines were all about how MongoDB was “well positioned” to gain from AI applications.

(The NoSQL company ended the quarter with over 43,100 customers, adding 2,300 new logos during the quarter; nearly 200 per week.)

MongoDB AI opportunities: MindsDB combo looks good.

Ittycheria didn’t deny the MongoDB AI opportunity however and went back to first principles to explain some of MongoDB’s fit as AI takes off.

“Results that come from training and LLM against content are known as vector embeddings. Content is assigned vectors and the vectors are stored in a database. These databases then facilitate searches when users query large language models… The key point is that you still need an operational data store to store the actual data. There are some adjunct solutions out there that have come out that are bespoke solutions but are not tied to actually where the data resides, so it's not the best developer experience..."

It was not immediately clear what adjunct solutions was referring to, but one MongoDB itself has espoused as useful (that does, indeed, not actually sit where the data resides) is MindsDB; more below.

See also: Natwest goes from zero to 900m API calls, eyes more data innovations, better dev journeys

"We believe AI will be the next frontier of development productivity; developer productivity and will likely lead to a step-function increase in software development velocity. We know that most organizations have a huge backlog of projects they would like to take on, but they just don't have the development capacity to pursue. As developer productivity meaningfully improves, companies can dramatically increase their software ambitions and rapidly launch many more applications to transform their business" said Ittycheria on June 1, 2023.

“It's still very early days. I think people tend to overestimate the impact of new trends in the short term but underestimate them in the long term. I think you're going to see a lot of things happening over the course of the next few months and quarters and years, but we feel we're in a very good position to take advantage of this new trend,” MongoDB’s CEO added.

MongoDB Q1 earnings shine but firm still loss-making

He was talking to analysts on an earnings call as the NoSQL database SaaS provider as revenue for the quarter (fiscal Q1 2024) rose 29% year-on-year to $368.28 million, beating estimates by $21.24 million. MongoDB is still loss-making, but trimmed quarterly losses from $77.3 to $54.2 million.

When it comes to turning MongoDB into a predictive database, the company's engineers have earlier espoused the benefit of working closely with MindsDB, a popular open source project that brings machine learning into databases by employing the concept of AI tables; machine learning models stored as virtual tables inside a database. (See details here.)

MindsDB is not a rival database, despite the "DB" in its name, but a service for integrating machine learning into databases that avoids the need for complex hands-on work with framework like Pytorch or Tensorflow to develop machine learning models. (MindsDB this week raised $25 million in a new round led by investor Mayfield, along with participation from TQ Ventures and existing investor Benchmark Capital as interest grows.)

As a MongoDB team earlier noted in a 2021 blog: “The MindsDB AutoML framework extends beyond most conventional automated systems of hyper parameter tuning and enables novel upstream automation of data cleaning, data pre-processing, and feature engineering.

"To empower users with transparent development, the framework encompasses explainability tools, enables processing for complex data types (NLP, time series, language modeling, and anomaly detection), and gives users customizability by allowing imported models of their choice."

“MindsDB also generates predictions at the data layer—an additional, significant advancement that accelerates development speed. Generating predictions directly in MongoDB Atlas with MindsDB AI Tables gives you the ability to consume predictions as regular data, query these predictions, and accelerate development speed by simplifying deployment work-flows" they added. Expect to hear more about this synergy in coming months.