Co-Founding FutureRobot as CTO


I am very excited to announce that I have co-founded a robotics company as CTO. We aren't announcing what we are doing yet, but you can keep track at futurerobot.co

I have had a passion for building new products for as long as I can remember. My passion for building robots has been here for even longer. I can remember in 7th grade implementing a convolution in C++ and also later in JAVA (C++ was infinitely faster...) from the ground up because I wanted to build a face detector and the paper showed how to do it. It ran on the Pentium computer I had with a 320x240 analog capture card and I think I got about 1 frame every second or two.

I was lucky enough to go to MIT and get to work in Cynthia Breazeal's Robotic Life group in the MIT Media Lab. I had read all her papers in high school and built 2/3 of a full robotic head as an extension of my computer vision project. It was all based on Kismet, so it was extremely exciting to get to work on Leonardo.

Kismet was the robot which made me into a roboticist.

When we were building Leonardo, we had a dedicated gigabit ethernet network, at least 8 full workstations which all had to be powered on and then huge racks of actuators, motor controllers and other things. I would guess the "robot" weighted about 10 pounds, but the secondary hardware weighed at least 300 lbs.

What you don't see when looking at Leonardo are the hundreds of pounds of computers and hundreds of meters of network cables.

From these experiences, and my learnings at EmSense about scaling complex sensors and large data, I found that the majority of time and challenge is not build a robot or implementing the algorithms for it. 

It is "everything else".

Our new startup's goal is to fix that for the next generation of robotic developers and users. Stay Tuned.