MIT Startup Exchange Feature:

Daniel de Wolff

MIT Startup Exchange’s founder and CEO George Shaw started thinking about use cases for anonymous tracking data when he was a researcher in Professor Deb Roy’s Cognitive Machines Group at the MIT Media Lab. He had yet to coin the term “spatial intelligence,” thereby defining a field, but Shaw and his colleagues were already exploring the approach that would lead to the founding of “I leveraged the knowledge of the world’s foremost experts in computer vision and person tracking at MIT, and in so doing, my focus shifted to understanding the tracking–not generating the tracking, but actually understanding it–building statistics around it, that could lead to actionable business insights,” says Shaw.’s spatial intelligence platform solves a problem that has long affected the industry: management’s ability to understand and assess how essential interactions within an organization’s physical spaces are relevant to its success—interactions between customers and the physical space, customers and store personnel, customers and other patrons, and even between employees and robots.

Prior to founding, Shaw held senior technical positions at Intel and Target, working respectively as a Principal Engineer and Principal Data Strategist in Consumer IoT. He also led research and development for a startup called RetailNext, eventually becoming CTO of the organization while helping grow the business from 10 to 250 people. By the time Shaw left, he had built groundbreaking technology focused on understanding people in retail stores, and the company had gone from functionally zero in revenue to about $US 45 million in annual recurring revenue.

Today, Shaw leads a team of thought leaders in machine learning analytics with the experience to provide advanced technology solutions to brick-and-mortar retailers and other industries with physical footprints. And the advisory board is stacked with esteemed MIT alums like Rony Kubat, who, in addition to being Shaw’s former lab-mate at the Institute, is the CTO for MIT STEX25 startup Tulip Interfaces. Also on hand is former Head of Product for Uber Birju Shah, a graduate of the Sloan School of Management at MIT.

“The key is separating the tracking from what we do with the tracking,” says Shaw.’s proprietary software integrates feeds from existing infrastructure to anonymously capture human behavior and interaction. The software has been re-architected as an ecosystem enabler, providing the flexibility to run any computer vision model, whether it is created in-house, from open source, or third-party partners. To be clear, Shaw isn’t looking to reinvent the wheel when it comes to AI. “We use commodity computer vision very consciously, but we are focused on breaking new ground in behavior analytics,” he explains.

And privacy is baked into the solution. The tagline is, “You can learn a lot from a dot,” because doesn’t generate any data about people beyond location: no facial recognition, no age or gender identification, no re-identification when re-entering a store, no personally identifiable information period. The result is a powerful tool for understanding human behavior and interactions that is GDPR and CCPA compliant right out of the box. And the stack runs on a local server, so everything stays off the cloud and on premises.

Shaw thinks about partnerships for his startup in two dimensions. On the technology side, there are camera manufacturers, network video recorder providers, and video management system providers. All of these companies want to make their hard drives more valuable by running analytics, which is where steps in.retail.

The other dimension is go-to-market. “There’s a real hunger for what we’re offering in the retail space—particularly grocery, big-box, and luxury retail,” says Shaw. With’s platform, customers can implement what Shaw refers to as “playbooks” catered to their specific needs. Two big-box retailers might have similar store footprints, but they could have very different goals. One might be interested in the ratio of the length of its queue to the number of open checkouts. Another might focus on staffing optimization. provides the necessary insights to dynamically address both of these issues and many more.

The MIT-connected startup opens up a whole new world of algorithm-based intelligence and actionable insights that can help guide the live interactions that matter most to a business as they take place. To expand their capabilities, Shaw and his team have recently partnered with the Loss Prevention Research Council at the University of Florida on a research and development project that aims to understand suspicious behavior and shoplifting. “We’re at a point where we think we can identify shoplifting behavior as it’s happening, in real-time, with nothing but the dot on the map, so there’s no risk of bias whatsoever,” he says. “If we can solve that problem robustly, it’s going to be a very large business for us.”

Aside from a host of pilots with big-name retailers, the solution is being deployed in commercial real estate, where understanding building lobby traffic is of interest to owners who want to optimize the use of on-site guards while understanding customer wait-times for amenities like the fitness center or the elevator. Manufacturing also stands to benefit from’s analytics when it comes to optimizing routes and exploring safety use cases around autonomous robot-human interactions.

As for joining the newest cohort of MIT STEX25 startups, Shaw is well aware of the opportunity it presents as far as achieving his goal of scaling to 1,000 stores by the end of the year. “I’ve seen the benefit STEX25 has had for friends who have gone through the program,” he says. “We’re now primed for access to some of the biggest and best companies in the world. We’re honored, and we’re already seeing the benefits.”