Rockfish helps businesses leverage synthetic data

For years, Vyas Sekar would call his old friend Muckai Girish from his undergraduate course to discuss potential startup ideas and get Girish’s input. The two usually talked about an idea and ended the conversation with it. When Sekar called Girish in early 2022 with an idea involving synthetic data, the conversation didn’t just end when he hung up.

Sekar and colleague Giulia Fanti of Carnegie Mellon University have been working on creating synthetic data to overcome the replication crisis, or data reproducibility, within academia. While Sekar saw the need for a solution in academia, Girish knew then that his clients were facing the same problem. After talking with several enterprises, the thesis was further confirmed.

“At the time, it felt like it was very real and there was an opportunity,” CEO Girish told TechCrunch. “So that got us going, and over the next few months we talked to some investors, people we knew, and more importantly businesses, and we realized that this was a significant challenge and worth putting a whole life behind. “

The result was Rockfish, a startup that uses generative artificial intelligence to create synthetic data for operational workflows to help businesses break down data silos. Rockfish integrates with database providers including AWS and Azure, among others, and helps users choose the best configuration for their data based on company policies or data usage.

Synthetic data is becoming an increasingly hot topic in the world of artificial intelligence, but when the company launched in June 2022, there was already growing momentum for it. Girish said he wanted to make sure Rockfish created a product that stood out from its own. peers and also a solution businesses will use every day, not every time.

Therefore, the company’s product is designed to continuously ingest data and is focused on operational data, including data such as financial transactions, cyber security and supply chains. These fields are constantly producing information for companies and are constantly changing. Focusing here helps Rockfish stand out from its competitors, Girish believes.

The company now works with several enterprise customers, including streaming analytics platform Conviva, in addition to government departments including the U.S. Army and the U.S. Department of Defense, Girish said.

Rockfish is announcing a $4 million seed round led by Emergent Ventures with participation from Foster Ventures, TEN13 and Dallas VC among others. This brings the company’s total funding to approximately $6 million.

Anupam Rastogi, managing partner at Emergent Ventures, told TechCrunch that he followed Sekar long before Rockfish was founded. He said what led the firm to invest was “the team, the market and the product, in that order.” Plus, Rockfish’s focus on building for enterprises made it a better fit for Emergent than some of the other players in the space.

“The team is super high-quality data scientists, multiple PhDs,” Rastogi said. “It’s a space that we think is very technically complex, and having technical power around the table is really, really important. “They have done a lot of groundbreaking work in the space, not just in the company, but in the entire industry.”

While Rockfish hopes its focus will help give it a moat among competitors, that doesn’t change the fact that synthetic data will be an increasingly crowded market. Artificial intelligence companies are turning to synthetic data, as many players think the market has exhausted other AI training data.

There are already numerous startups looking to tackle the market, including AI tonicRaised over $45 million in venture funding; Mainly AIwhich raised $31 million in VC funding; and Mistyraking in $14.5 million before being acquired by SAS in 2024, just to name a few.

The company wants to expand its approach to synthetic data by including other types of models, such as state-space models and mathematical models that use state variables, Girish said. The company is also working on improving its end-to-end features.

“It’s not like taking random data and creating synthetic data for the web,” Girish said. “There is no guarantee that it will be good. But if you put all of this together for businesses, it’s actually very relevant and real. So that’s the key to that, and then being able to do that permanently is what we find useful.