Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > EDA/IP

Dynamic duo creates 'capital efficient' startup

Posted: 01 Dec 2004 ?? ?Print Version ?Bookmark and Share

Keywords:eda? vendor? president? ceo? cto?

Madhani: What attracts geniuses is that you give them an unbelievable problem and make believers out of them.

Taking on big EDA vendors with a new physical-synthesis product is a huge challenge, but Pravin Madhani, president and CEO of sierra Design Automation, makes it sound like the most natural thing in the world. The ease and confidence Madhani exudes have a lot to do with the engineering savvy of his CTO, Shankar Krishnamoorthy, who formerly headed r&d operations for the EDA industry's most popular synthesis tools.

Sierra made waves in April when it announced Pinnacle, a physical-synthesis tool that promises to handle 10Mgates flat in an overnight run. It is a bold and audacious claim for a tiny startup that is going up against well-established players.

But Sierra has more credibility than the average EDA startup. It is not because of a massive infusion of venture capital, and Madhani insists on running a "capital-efficient" operation with a small staff and a high level of automation.

Starting with nothing more than an idea in January 2003, Sierra now employs 23 people, 22 of them engineers. The company is shipping Pinnacle and is earning "good revenues," according to Madhani.

Growing up in Mumbai, Madhani knew early on that he wanted to start his own company. But first, tackling a personal challenge, he passed the entrance exam to the prestigious Indian Institute of Technology (IIT). There, he studied VLSI design before heading to the University of Texas in Austin to earn his master's degree in 1991.

"Because my family was all in business, the immediate thing was to go start a business, but you can't just go out and do that here," Madhani said. So he took a job at Silicon Graphics Inc., where he wrote a timing-analysis tool. Then he tried to start a timing-analysis company, but the venture capitalists he approached turned him down, saying the potential market was too small.

Krishnamoorthy: The advantage of building from scratch is that you can use the latest tools and techniques to get an order-of-magnitude improvement in efficiency.

Meanwhile, as Madhani was growing up with dreams of entrepreneurship, Krishnamoorthy, who hails from the south of India, had his sights set on engineering. "My dad was a computer professional, and I was fairly aware of what was happening in the computer field, so when I passed my exams at IIT, I went into computer science and engineering," he said.

Krishnamoorthy also attended both IIT Mumbai and the University of Texas at Austin, one year behind Madhani, and the two men knew each other in passing. After graduating in 1990, Krishnamoorthy went to work at Synopsys as an R&D engineer. He headed the Design Compiler R&D team in 1995 and 1996, at a time when Synopsys was fending off competition from startup Ambit. Then he started, and ran, the PhysOpt R&D team that produced Physical Compiler.

Madhani said the idea for Sierra came out of discussions he and Krishnamoorthy had been having with some people in the industry, talking about capacity challenges of 130nm and 90nm designs. It all crystallized a year ago in January, he said, and by February 2003 the two had raised $5 million and were starting to recruit other engineers.

Building a startup
For Krishnamoorthy, who was used to the extensive infrastructure available at Synopsys, those first days at Sierra were somewhat of a shock. "When I walked into an empty office with wires hanging off the walls, it felt like a huge challenge," he said. "But the advantage of building from scratch is that you can use the latest tools and techniques to get an order-of-magnitude improvement in efficiency."

Madhani, who had been through it all before, said his second startup was easier than the first. "I already had pre-established relationships, so putting things together was pretty quick," he said. "It's much easier to walk into customers and talk to them" the second time around.

He leveraged an extensive database to hire top-notch EDA engineers away from companies like Cadence, Synopsys and Intel. "Basically, we got all the best people," he said. "We told them we had a big problemhow to do 10Mgates overnight, flat. Some didn't believe us. What attracts geniuses is that you give them an unbelievable problem and make believers out of them."

Next challenge: to run this team so efficiently that it can compete with big EDA vendors that employ thousandsand which can, and will, muster major efforts to stop upstart competitors like Sierra.

Madhani, who believes the best EDA startups haven't had massive venture capital infusions, isn't worried about finding money. "We have good revenues right now," he said. "Between that and our liquidity, we don't need to raise a lot of money." Sierra has a handful of job openings but is looking to grow "organically," he said.

But Madhani does anticipate growth, and an eventual IPO. IC implementation, he noted, is the biggest single market segment in EDA. "Clearly this segment is so large we can grow this into a large company, and then build upon that to do other things," he said.

- Richard Goering
EE Times

Article Comments - Dynamic duo creates 'capital efficie...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top