Former Googlers Launch Startup To Address the 'Hard Problems' of Enterprise AI/ML

RealityEngines.AI, the AI and machine learning research startup founded by former Google executives and engineers, came out of stealth mode today and unveiled its flagship products and its mission: To make machine learning (ML) easier for the enterprise.

Specifically, the company engages in research and offers cloud AI services "to help companies of all sizes to easily and effortlessly embed cutting-edge deep learning models into their business processes or customer experiences" the Web site states.

The San Francisco-based company's first product offering is a set of tools designed specifically for some standard enterprise ML use cases—things like user churn predictions, fraud detection, sales lead forecasting, security threat detection, and cloud-spend optimization. But the tools can also be used to provide a more general predictive modeling service.

The company uses neural architecture search (NAS) and generative adversarial networks (GANs) to "take you're noisy/incomplete data and create an accurate deep learning system without any manual tweaking or data science." GANs enable the generation of synthetic data that can be combined with an organization's real data to create accurate deep learning models.

The company is emphasizing the research it is actively pursuing, and its focus on a list of "hard problems" enterprises implementing AI/ML technologies are facing, including meta-data learning, combining neural nets with logic rules/specifications, and transfer of learning. Details of the company's efforts are posted on its Research page.

There's a demo on the company Web site showing off its tech. Visitors can upload a picture of their faces and generate different emotion and change their ages, all with a click.

CEO and co-founder Bindu Reddy was GM for AI verticals at AWS. She previously served as head of product for Google Apps. She founded the company with two other former Googlers: Arvind Sundararajan and Siddhartha Naidu. Sundararajan, who is the CTO, was most recently a senior technical lead at Uber's autonomous vehicles division. During his time at Google, he was the technical lead for AdSense's machine learning and serving system, and the technical lead for Gmail's backend system. Naidu, who is the director of research, was most recently a principal engineer in Amazon's supply chain organization. He spent nearly a decade at Google, where he worked on large-scale machine learning projects. The company's list of investors includes former Google CEO Eric Schmidt.

About the Author

John K. Waters is the editor in chief of a number of sites, with a focus on high-end development, AI and future tech. He's been writing about cutting-edge technologies and culture of Silicon Valley for more than two decades, and he's written more than a dozen books. He also co-scripted the documentary film Silicon Valley: A 100 Year Renaissance, which aired on PBS.  He can be reached at