We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!

AI startup Abacus.ai today announced the close of a $13 million series A round for its cloud platform for AI services. The company was founded last year by a team with experience scaling AI projects at AWS and Google, including BigQuery cocreator Siddartha Naidu, former AWS manager of AI vertical services Bindu Reddy, and former Uber engineer Arvind Sundararajan. The company also announced today that it’s changing its name from Reality Engines to Abacus.ai.

Reddy, who serves as CEO, and Sundararajan previously cofounded Post Intelligence, a startup Uber acquired in 2017.

“We think machine learning and deep learning is as much about infrastructure and systems as it is about algorithms, and both Arvind and Siddartha are really good in terms of building large-scale, real-time, big data systems. And I think that’s really become very useful for us,” Reddy told VentureBeat in a phone interview.

The round was led by Index Ventures, with participation from individuals like former Google CEO Eric Schmidt and Ram Shriram, a founding board member at Google. Schmidt and Shriram funded a seed round for the startup last year.


Transform 2022

Join us at the leading event on applied AI for enterprise business and technology decision makers in-person July 19 and virtually from July 20-28.

Register Here

Abacus.ai helps businesses put models into production-scale AI strategies. A number of business executive surveys have found that beyond tech giants with a lot of AI talent, like Facebook and Google, many businesses struggle to implement AI beyond a series of minimally viable products.

After evaluating a user’s data set, the company decides which neural network architecture is best for a particular use case through a neural architecture search (NAS) algorithm. Abacus.ai researchers detailed their NAS approach in a paper published last fall.

Abacus.ai also uses synthetic data produced by generative adversarial networks (GANs) for training models. Other AI service companies helping enterprise customers deploy AI using synthetic data include Element AI and Andrew Ng’s Landing.ai.

The funding will be used to improve existing Abacus.ai services, grow the company’s customer base and research team, and expand into language model and computer vision services. The company primarily works with columnar data today.

As part of the round, Abacus.ai is launching its AI services out of beta, including the model showcase service that lets customers quickly share or compare deep learning models. The company says it worked with 1,200 testers for the beta phase.

Abacus.ai was founded in March 2019. and has raised a total of $18 million. It’s based in San Francisco and currently has 22 employees.

VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Learn more about membership.