Founder Spotlight: Realtime Robotics
From four founders with a prototype to Series A and transforming robotics and automation.
By: Ryan Mendoza
Welcome to the first installment of Scrum Venture’s “Founder Spotlight” series. As an early-stage VC firm bridging U.S. innovation hubs, such as Silicon Valley, and leading Japanese corporations, we have the pleasure of working with transformative startups — from consumer to enterprise sectors of all types — from the very start. In this series, we’ll do a deeper dive into our portfolio from the founder perspective — from the ideation phase, major milestones and all those entrepreneurial lessons learned along the way.
This week, on the heels of their $11.7M Series A funding round, we’ll chat with the founding team of Boston-based startup, Realtime Robotics — the inventor of responsive motion planning for industrial robots and autonomous vehicles.
Emerging technologies in automation have incredible potential to transform industries across the board — but especially in manufacturing, agriculture, food service, construction, healthcare, and consumer settings. We first invested in Realtime Robotics Seed round in 2017 and are proud to continue to support its journey with the latest Series A and beyond.
We asked Founder & Chief Architect Dan Sorin and Founder & Chief Roboticist George Konidaris a few questions about their experiences and the Realtime Robotics journey — from “four founders, an academic-grade prototype, and a patent application” to Series A and transforming robotics and automation.
Tell us more about Realtime Robotics. What’s the company’s mission?
Dan Sorin: Realtime Robotics was founded with the goal of transforming how robots and autonomous vehicles move. Our initial invention was a proprietary computer processor that quickly solved how to get a robot or vehicle to its desired target without collisions. This solved the problem of conventional motion planning, which is too slow for robot and AV applications.
Since this initial innovation, Realtime has continued to transform automation, with developing products that will provide trailblazing features like risk-aware driving, high-productivity multi-robot workcells, and automated robot vision that continuously calibrates itself. Automation offers enormous opportunities, and Realtime continues to develop products that achieve more and more of that potential.
We started Realtime Robotics in March 2016 at Duke University with top researchers Sean Murray and Will Floyd-Jones. We are both professors, George being in Robotics and Motion Planning and myself in Computer Architecture. Since its inception, the Realtime team has grown to 40 people at its Boston headquarters, under the leadership of CEO Peter Howard.
Can you tell us a little more about your backgrounds? What were you doing before starting the company?
Dan Sorin (DS): I’m a Professor of Electrical and Computer Engineering at Duke University, and my research has focused on computer architecture. I’ve always been interested in designing computer hardware, either for performance, reliability, or verifiability. I’ve also sought out collaborations across disciplines, including projects with researchers in formal verification, information theory and robotics. Prior to joining the faculty at Duke, I earned my PhD at the University of Wisconsin, and before that, I was an undergraduate at Duke, where I first got hooked on computer architecture. While my career has been in academia, I’ve had longstanding interactions with the computing industry, including a 6-month sabbatical at Intel.
George Konidaris (GK): I was born in Johannesburg, South Africa, which is where I did my undergraduate degree at the University of the Witwatersrand. I then received a scholarship to study in the UK and completed an MSc in Artificial Intelligence at the University of Edinburgh, before coming to the U.S. to do my PhD at the University of Massachusetts Amherst. After a postdoc at MIT, I joined the faculty at Duke, which is where I met Dan and started Realtime Robotics.
Let’s talk inspiration — what was your “ah-ha” moment when you first realized you wanted to start the company?
DS: For me, the moment was when I saw our first demo at Duke. I realized that it had not taken very long for only two professors and two students (the founding team) to create a first prototype that outperformed the industrial and academic state-of-the-art motion planning solutions by more than 1000x.
We were already so far ahead of what everyone else was doing, and there were such clear commercial opportunities. But we realized that for this technology to have a real impact, we had to be the ones to bring it to market.
Tell us about the culture at Realtime Robotics — what makes you different from other startups?
DS: Our down-to-earth team has equal parts: those who are pushing technology towards the future and those who have an expectation for delivering a reliable, high-performing product. It is rare for companies at any stage to have this kind of combination and high-level collaboration that our team does.
We each value input and trust each other to do the work. And I think this is what our customers appreciate about working with us. We are honest and fair with what we can achieve so that our products and partnerships can be both competitive and successful.
What were some of the biggest challenges you have had to overcome as a founder?
DS: This is probably common to most tech startups, but the gap between exciting new technology and a marketable product has been wider than initially expected. Among other issues, the “cool” technology has to be easily usable and deployable by customers, and it has to be more robust than a demo that can occasionally fail.
What one piece of advice would you give to other entrepreneurs looking to get their company off the ground and secure funding?
DS: As an academic researcher, I’d recommend to others in academia that they find business leadership they trust and not assume that they are going to be able to quickly become the business experts their companies need.
What kind of momentum has the company experienced thus far? What kind of milestones have you reached?
DS: The momentum of Realtime is perhaps the most thrilling part of this whole experience. We started with four founders, an academic-grade prototype, and a patent application. From there, we slowly added engineers, business leadership, and patent applications, and our demos became more sophisticated. Then there was this acceleration period during which the teams grew, we added salespeople, and started working on proof-of-concept prototypes with customers and partners.
I’ll never forget the first time I noticed an employee whom I didn’t know. I had to ask George who this person was, which seemed like an inconceivable situation to have reached after starting with just four of us. And it just continues to accelerate!
The release of our first commercial solutions (RapidPlan and RapidSense) earlier this year was a significant milestone for us — — you can see the automatic planning technology in action here. We are excited to continue to focus on developing our solutions that expand the potential of robotic automation.
Speaking of milestones, congratulations on the Series A — What’s Next? What do you plan to do with the new funding?
DS: This round of funding will enable Realtime to move from an early, limited-release product and proof-of-concept projects to large-scale commercial products. We have already proven the technology and this round will enable us to prove the products based on it.
Why did you choose to partner with Scrum Ventures? Where are you hoping the team will provide most value?
DS: Scrum Ventures was a valuable partner on our seed round, and we’re delighted to continue the partnership. Beyond financial support, Scrum has helped us with introductions to other VCs and to companies that are potential customers and commercial partners. Scrum has also helped us improve our messaging and presentation of the company while sharing useful insight into the robotics and autonomous vehicle markets.