How do you go from working for Bell Labs to starting up a number of EDA companies? Atrenta's Ajoy Bose tells of his transitions from the largest to the smallest of companies and what he learned along the way...
A couple of weeks ago I had the opportunity to sit down with an industry CEO who normally does not like to be in the limelight. I am talking about Atrenta’s Chairman, president and CEO – Dr Ajoy Bose. We started at the very beginning of his journey into EDA and from there branched out into many different areas including mentoring, the state of the industry and advice for potential entrepreneurs. In this first blog we find out more about the person and his transitions from a very large company to an entrepreneur.
Brian: Why did you originally come to the US?
Ajoy Bose: I came here as a graduate student and did my masters and PhD at the University of Texas, Austin. There was a pocket of EDA at the university at that time. My PhD adviser was Professor Steve Szygenda, and he had created a commercial simulator called Tegas, which was probably one of the first or very early examples of commercial EDA. So I got exposed to EDA, I got exposed to simulation, and I liked it. In those days, electronic design was relatively primitive and tools were at their very early stages of development. It seemed like a pretty open field for research, for innovation, for building a career, so I decided to go with EDA very early on.
Brian: They were fun days, the very formative days of the EDA industry. Then you went to Bell Labs, right?
Ajoy Bose: That’s right. Most of what I did at Bell Labs was EDA related. Design and EDA, in those days, were organizationally very close. You used to report into the same boss and people would rotate between doing EDA and doing design. You would work in an office where a neighbor in your office could be using your tools and I think of that as one of the best days of EDA innovation, because of this close coupling between designers and the tool developers.
Brian: You then moved from Bell Labs into Cadence.
Ajoy Bose: Yes. It’s interesting. I mean, this is AT&T, this is Ma Bell. It’s kind of an ecosystem in itself. It was in a very comfortable, well-contained, very mature environment. Going from that to my first exposure to Silicon Valley was a big change. The Bell System, at its peak, was like a million people. They had started restructuring while I was still there, so it wasn’t a million people when I left, but that’s the kind of environment I was used to. Going from that to a company that had a few thousand people was a big transition. After that I went from a company with a few thousand people to just a few people and that was not as traumatic.
Brian: What was it that made it traumatic for you?
Ajoy Bose: It was more in my head than anything else. It was that comfort and the safety net that you take for granted when you were in that big environment. I had that feeling that the safety net was gone.
Brian: Presumably the other side of it is that you can have a real impact.
Ajoy Bose: That was the new learning. It was a different world. I missed what was good in the old world. Then I started seeing the good in the new world, and I learned a lot from it. The ability to have an impact, the ability to do things with the objective of generating business success, which I think is a good thing. That grew on me to a point where it tipped me to becoming an entrepreneur myself. Before, I would do things for the scientific or the engineering success. I learned that in EDA you are dealing with a very distributed user community. Here, you have a very heterogeneous group of users and you have to satisfy all of them. These are interesting problems.
Brian: And what were you working on at Cadence?
Ajoy Bose: I was supposed to join Gateway as the VP of Engineering. By the time I got to Gateway, they had become Cadence, so I effectively ended up as a VP of Engineering for the Cadence front end tools, so it was Verilog, VHDL simulation, synthesis, timing, test. It was a fairly large chunk of Cadence’s products at that time.
Brian: So how long were you there before the entrepreneur bug bit?
Ajoy Bose: I was there for about two and a half years. I didn’t have any desire to start my own company when I was in Bell Labs, but Cadence gave me that first taste of blood and it also gave me the confidence that, “hey, you know, maybe it’s not that big a deal. I can start my own company”.
Brian: Was it a technical idea that you had or was it just “I want to start a company and then I’ll work out what I want to do”?
Ajoy Bose: It was an interesting idea. I started the company with another gentleman, who had been an entrepreneur before me, so I had somebody who knew what he was doing. Cadence had a team in India that used to report in to me and this gave me my first taste of offshore engineering. The idea was that we could do services based in India with a focus on software quality and software processes. This was before doing offshoring and services companies had become popular.
I did that for a few years, but it wasn’t completely my company. It was a shared experience, which was very helpful and I’m very grateful to my partner. My idea was that I could start a services company like a multi-stage rocket, so I would fire the first stage that would provide my business, my company, I have some resources, I have the foundation on which I can fire my stage 2, which would take me on to some more specific, more focused areas. I started a company called Interra in ’95. Fairly soon I had a business that was generating cash, so I could experiment with the ideas. On top of this first stage, I went after several different business ideas, and today that has spawned several companies. So I transformed from a serial entrepreneur to a parallel entrepreneur for a while.
Brian: And was Atrenta one of those experiments?
Ajoy Bose: Yes. Atrenta was one of those experiments. Atrenta came out of a services project that we did for a large semiconductor company that was looking for a reuse checker. They had defined a set of criteria that a design had to pass to be considered reusable. They wanted somebody to build a tool to check designs for those properties. So we did that, and we had worked out an arrangement with them that we would retain the rights to the technology. After we completed that project, I realized that this could become a stand-alone, interesting EDA product company.
Brian: And you continued to run both companies?
Ajoy Bose: No, I stopped at that time. I used to run like three companies in parallel, but when I created Atrenta, I stopped running the other companies. I’m still involved with them on the board, but I focused my operational attention fully onto Atrenta. We started out by looking at a design from its reuse properties, but we had created a framework, an architecture on which you could look at design for its test properties, its power properties, its clock properties. This whole notion of looking at a design from these different perspectives, at the very beginning of the design cycle, and being able to help a designer to identify and fix problems seemed to me like a pretty cool idea. I said I can make a company out of this and that became Atrenta.
In the second part of this interview series we will examine the state of EDA and the future for Atrenta.