Seattle company's silicon chips make communications products efficient

Who: William Colleran

What he does: Chief executive officer of Impinj

What it does: Founded by renowned chip expert Carver Mead and Chris Diorio, a University of Washington associate professor of computer science and engineering, the Seattle company invented "self-adaptive silicon" chips that make communications products smaller, faster and consume less power. In September, MIT Technology Review named Impinj one of "the most original and promising companies formed in the last few years."

Résumé: Colleran, 40, is former director of business development for semiconductor designer Broadcom of Irvine, Calif.; CEO of Innovent System, later sold to Broadcom; investment banker with Merrill Lynch; senior staff engineer with TRW. He holds a Ph.D. in electrical engineering from UCLA and a law degree from Harvard Law School.

Learned chips: "Self-adaptive silicon enables us to change the way in which transistors function after they've been manufactured. When the chips are being made, the functionality is locked in place. That's the end of it. Our technology adapts those transistors so they can have optimum performance once they've been deployed. They self-improve, or adapt their performance. In a specific product, you would be able to deliver the same functionality, but with one-tenth of the power, because the transistors are optimizing themselves to not be wasteful."

Coming to a tower near you: Cellular towers have within them electronics for both transmitting and receiving radio signals. The chip should allow a bay station to transmit signals more reliably and with less power. Cellphone companies "can put up fewer bay stations or handle more cellphone calls with the same number of bay stations," he said.

No revenue — yet: The company today is launching its first commercial product: tools that allow a communications-system developer to integrate the self-adaptive chip into its own products. Colleran said to expect modest revenues this year and "more meaningful numbers" next year.

Funding: $15 million from Arch Venture Partners and Madrona Venture Group in September 2000. It should close a second round soon, bringing in one more, yet-to-be named investor.

Five years out? "If we're able to execute on our plan, and if the markets cooperate, we could clearly be a public company in some very high-growth markets."

— Monica Soto