Moore's Lobby Podcast

JITX Aims to Provide Superpowers to Circuit Design Engineers

Episode #58 / 1:00:35 / December 20, 2022 by Daniel Bogdanoff
0:00
Technical Sponsor: ROHM Semiconductor

How do you begin by building world-record robots and end up trying to revolutionize the electronics industry? Dr. Duncan Haldane, co-founder and CEO of JITX, retraces his fascinating career path and describes their vision for a better circuit design process.

Duncan Haldane has always approached problems a little differently. In high school, he scavenged auto parts to build robots. In college, Haldane was mimicking biology to create two Guinness world-record robots: the speedy, slightly disturbing X2-VelociRoACH (video) and the crazy, jumping Salto (video). 

X2-VelociRoACH: world’s fastest robot relative to size. Image used courtesy of University of California, Berkeley

Now he is leading JITX as it builds a new design flow process to automate circuit board design. 

 

I think the motivation is, how do you take the best of how we as a species know how to design hardware and apply it to every single design? How do you make the world's most expert knowledge reusable, repeatable, scalable, in the way that software is?

 

They generate designs with code rather than schematics and use AI to help automate the flow and improve rule checking.

Screen capture of the JITX design environment. Image used courtesy of JITX

According to Haldane, JITX software is targeted at senior engineers to make them more productive. The goal is to automate routine tasks like pin assignments, design checks, and even help users’ identify suitable parts that are in stock at distributors. 

Haldane recounts how they began designing for customers using early versions of their tool and kept improving by “noticing what didn't work and what was really bad.” He added that they are “making a tool for experts,” so it must “meaningfully help them very quickly, or it's not worth anything at all.”

Other highlights from this Moore’s Lobby interview of Duncan Haldane include:

  • Why “robots suck pretty bad as compared to animals”
  • How simple errors in mass distribution calculations can result in robots that “spin like a maniac”
  • Discussion of the data challenges that circuit and PCB designers still face
  • Advice for tech entrepreneurs who are pursuing venture funding

 

A Big Thank You to This Episode's Technical Sponsor

 

 
 

Meet Duncan Haldane

Duncan started working on JITX because he wanted a better way to design robots. While he was at Berkeley, he started collaborating with JITX co-founder Jonathan Bachrach on better design tools for origami robots. Since co-founding JITX in 2017, he has been working on sales, marketing, strategy, fundraising, and recruiting.

 

 

Duncan holds a bachelor’s degree in mechanical engineering from Florida State University and a master’s and Ph.D. in mechanical engineering from the University of California at Berkeley, where he explored new ideas in the field of bio-inspired robotics. Duncan's work in robotics won him a best-paper award from the IEEE International Conference on Intelligent Robots and Systems. In addition, a set of his robots became de facto scientific platforms for an international community of robotics researchers.