Larry Carin, PhD of Infinia ML

October 12, 2020April 7, 2023 thetechtribune

An exclusive Tech Tribune Q&A with Larry Carin, PhD, the founder and Chief Scientist of Infinia ML, which was honored in our:

Tell us the origin story of Infinia ML – what problem were you trying to solve and why?

The idea for our Infinia ML began in my lab at Duke University. Companies were coming to us for machine learning help, and we decided to spin out a new company to meet what we saw as a growing need.

Lots of people focus on the theoretical benefits of the most advanced techniques. While we have those capabilities, that was never our focus. Our focus was always on the application of the technology to real business challenges.

What was the biggest hurdle you encountered in your journey?

Infinia ML started by providing custom ML solutions for a wide variety of companies, challenges, and data types. Our expertise in machine learning is deep but also very broad. One challenge in these early years has been directing our focus to areas where we can add the highest value, which sometimes means not leveraging all of our breadth of expertise.

Another challenge is getting business leaders to understand that machine learning isn’t a standalone technology like some SaaS products they’ve probably implemented in recent years. ML needs to be customized to a company’s specific data and problem. It also requires ongoing attention and maintenance when the data around you starts to change.

What does the future hold for Infinia ML?

We are now focused on document analysis use cases as well as the auditing of ML models themselves. We are building software to aid both missions and make our team as efficient as possible.

What are your thoughts on the local tech startup scene in Durham?

We’ve got phenomenal technical and academic talent here in Durham and across the Triangle. Combine that with our quality of life, and I wouldn’t want to build a machine learning company anywhere else.

What’s your best advice for aspiring entrepreneurs?

I’ll give my advice for aspiring data scientists, which I suspect is broadly applicable: just because a tool is new, it doesn’t mean it’s the best tool for a particular job. There may be simpler, more proven methods for solving your problem that have been around for years. Find the best solution to the problem, not the most complex.