This is a very interesting conversation about technology from Stanford being commercialized in financial services and healthcare, addressing very specific use cases in risk and other domains.
Sramana Mitra: Let’s start by introducing our audience to yourself as well as to Ayasdi.
Gurjeet Singh: I’m one of the co-founders and Chairman of the Board at Ayasi. I’m a mathematician and a computer scientist. I grew up in Delhi, India. I ended up working at Texas Instruments as a chip designer for about a year. I made my way to the US to pursue graduate studies in Computational Math and ended up working across computational fluid dynamics and computational mechanics. I eventually ended up doing my Ph.D. in the Math department on topology.
Sramana Mitra: This is at Stanford?
Gurjeet Singh: Yes, correct. The genesis of that research was pretty interesting. Around the year 2000 or so, DARPA and NSF realized that the way people were doing science had changed. Since most of their research financing went into scientific endeavors, they realized that people started doing science by creating large complex datasets.
They felt that people who are constructing the best large complex datasets were likely not the best people to analyze those datasets and learn from them because they felt that if you’re trying to be biologist or a chemist, you may not have the best training in mathematics or statistics to be able to learn as much as you could. They had that idea. Computers and CPU’s kept getting cheaper. That was the research problem that they were interested in.
My Ph.D. advisor at Stanford thought of ways in which we would use this old branch of mathematics called topology and try to template this problem. During the course of my graduate work, we published papers on protein folding, cancer research, computer vision, and object recognition. When we were done with the research, DARPA essentially asked us to commercialize it. That’s how we came to start Ayasdi.
Sramana Mitra: What problem domain did you choose to focus? You talked about the horizontal side. What business problem were you going to solve?
Gurjeet Singh: We were doing something that, classically, people advise you not to do. The standard advice in entrepreneurship is find out a pain and then go solve the pain. We were in the situation where we had the technology and we didn’t really know the pain.
What made it even more challenging for us was we left Stanford with a bunch of research but the distance between product and research is humongous. When we first started the company, we spent the first three or four years without growing. We built prototypes of our research. These are things we believed could become products eventually.
Sramana Mitra: How did you manage to stay afloat during that period?
Gurjeet Singh: DARPA.
Sramana Mitra: The DARPA grant allowed you to finance the company?
Gurjeet Singh: We were an incorporated company. There is a route in DARPA called SBIR. We used the mechanism to found the company. We also did some work with other government agencies. You can think of it as consulting work, basically, for the first few years.
Sramana Mitra: When did you start the company?
Gurjeet Singh: We started the company in 2008. We didn’t really do a whole lot for two years. We officially started the company in 2010.