Automating With the Help of MSR

In September 2016, Rami Jachi founded Nousot, a Chicago-based company that is powered by artificial intelligence and deep learning to build predictive and descriptive models, and then update them without human intervention.

Nousot gathers and organizes data, identifies predictive variables, develops and validates models, and then refreshes them without human involvement.

Since founding the company, Jachi has built a connection with Northwestern's Master of Science in Robotics (MSR) program, and in fact, has hired two MSR graduates to full-time positions — one as a software engineer and one as a machine learning researcher.

Jachi recently took time to talk about his budding company and why he values being able to connect with the MSR program.

How would you describe what Nousot does to someone with no understanding of robotics or AI?

At Nousot, we solve complex problems using math. Artificial intelligence really has become a buzzword, but the thing with AI is it covers a host of very different applications. You need to be versatile enough to pivot between problems, but at the end of the day, the problems all require the same skill set. You need to have a deep understanding of applied mathematics.

How did Nousot come to be?

I was a quantitative data scientist focused on machine learning for the last 12 years. Along the way, I realized people were repeating work over and over, and that was a waste of talented people’s time. I figured that if we could automate some of that high-end, thinking work that was being repeated, then we could free up time for creative-minded people. Then, if their time is freed, they can tackle other problems.

In what ways have you been involved with the MSR program?

I know Todd Murphey very well; in fact, he’s on our Board. I’ve hired two MSR graduates, and I’m constantly looking at other students who are looking at more applied or more general ways to enter the workforce. MSR students have a massive amount of skills and creativity, and the program is extremely strong for that.

From your experience, what differentiates students who graduate from the MSR program?

One of the biggest things I think a lot of people don’t know is the students in Northwestern’s MSR program do a lot of machine learning. As a result, the problems that go into machine learning and AI are problems the MSR students are very well equipped for. Additionally, and I say this without exaggeration, the programming skills of Northwestern’s MSR students are second to none. I’ve also noticed that with MSR students, they have an attitude of what will it take to make things work, rather than a mindset of worrying about perfecting a model, which I think is more common to see out of more people with machine learning backgrounds.

What do you think is the biggest challenge facing the robotics industry today?

I think the biggest challenge is the hype surrounding robotics and the actual demands aren't perfectly aligned at this point. There is a ton of demand, but people are demanding it primarily for testing it out as opposed to baking it into their operations. That disconnect is true for any type of new, innovative technology.

Is there anything else you’d like to add?

I’m excited about the industry's next 10 years. It’s been a great ride so far.