Skip to main content
Mathematics and computation

Mathematics and computation

Finding a niche within scientific computing

13 Jul 2018 Margaret Harris
This article first appeared in the 2018 Physics World Focus on Computing under the headline "Finding a niche"

Tech-X co-founders John Cary and Svetlana Shasharina reflect on how the firm developed its specialism within scientific computing

Partnership: John Cary and Svetlana Shasharina. (Image courtesy: Tech-X)

Why did you decide to start Tech-X?

John Cary, chief scientist and CEO: When I was a professor of physics at the University of Colorado, and my wife Svetlana (Sveta) was a researcher there, we saw an opportunity for doing tech transfer within the Small Business Innovation Research (SBIR) programme. The US government created this programme to give small businesses a way to get research funding (which would otherwise all go to laboratories, big businesses or universities) and to help them translate that research into commercial products that could grow the economy.

What we did was to develop computer programs that take the fundamental laws of physics and use them to model how systems will behave and evolve. An example would be the klystrons found in particle accelerators. Klystrons are cavities – metal boxes or cylinders – where a beam of electrons enters at one end, and then a small amount of power is injected into the cavity to cause the beam to bunch. By the time the beam reaches a second cavity, it has begun oscillating, and that produces microwaves that can be extracted and used for a variety of purposes. Our software can tell equipment designers how these devices are going to work – how much power is going to come out, how narrow a spectrum, whether the frequency is stable or drifting, how much energy is wasted, and so on. That means they can try many, many configurations on the computer before they make the device, saving money and time.

How has the company changed over the years?

JC: At first, our growth was all based on the SBIR model. The US government would basically pay us to write the software that scientists at the Department of Energy use to build particle accelerators, and then we would turn around and sell that software to laboratories in other countries as well, such as the UK’s ISIS neutron source and facilities in Europe, China and Russia. During that period, the people we hired were mostly scientists and engineers who could write proposals for SBIR funding, and we also contracted with a company to help us understand the paperwork required to work with the federal government.

We developed computer programs that take the fundamental laws of physics and use them to model how systems will behave and evolve

More recently, though, we have started to attract customers from the commercial sector. We’ve worked with a firm called Applied Materials, for example, that builds large plasma devices for manufacturing integrated circuits, and we also work with companies in the aerospace, defence, and oil and gas industries. As we got more into commercial markets, we created a quality-control department, and we also brought in people who can help customers use our software, because it is fairly complex and you need to understand a lot of basic physics principles.

What have been your biggest challenges?

JC: Establishing a sales group has been tough. We have had some success in hiring scientists and engineers and teaching them how to sell, but we have not found the ideal senior person who combines skills in sales and in science or engineering.

Sveta Shasharina, senior scientist and co-founder: I can elaborate on that. Essentially, we tried to hire someone who could help us commercialize our product and find new customers, but we couldn’t find anyone who had a sales-type personality – driven and personable – who could also understand the technical aspects of the software well enough to run it and drive customer-oriented development. For the moment, John has taken on that role in addition to his responsibilities as CEO, but it takes a lot of his time, so we are still looking for that magical person.

What do you know now that you wish you’d known when you started?

SS: I used to think I knew how to communicate with people, but it has taken me several years to become relatively good at it. I’ve learned a lot since becoming a manager. I also wish I hadn’t spread myself so thin early on, because my curiosity led me into lots of projects that didn’t end up contributing to the company’s current direction. My background is in fusion plasma physics, but I was working on distributed computing, and now I’m splitting my time between R&D and the business or commercial side.

JC: When you start a company, you’re trying to find your way in the world. You’re figuring out what you can do. Then, as you grow, you realize where your biggest strengths are, and at that point it’s important to focus. That is hard, especially in a research-driven company where you have a lot of very smart people, and they have lots of ideas about how to go off in their own very smart directions. Running a business is different from managing a research group: it’s important to get people to focus and put their energies into a few areas, rather than doing whatever they want.

Any advice for someone starting a business in scientific computing?

JC: Figure out what your competitive advantage is and then test your theory. If you think your selling point is X, go out and say it a few times: “We are good because of X.” Then, if you don’t get any sales, go and think of an alternative. It is easy to get caught up in how wonderful your software is, but it won’t help unless you can communicate it to someone and they accept it.

  • Enjoy the rest of the 2018 Physics World Focus on Computing in our digital magazine or via the Physics World app for any iOS or Android smartphone or tablet.

Related events

Copyright © 2024 by IOP Publishing Ltd and individual contributors