Rob Devlin

The lens that gave up its curve

A camera lens has been a piece of bent glass since the 1600s. Rob Devlin's lenses are flat - patterned with billions of pillars smaller than the wavelength of light, etched on a wafer in a chip foundry.

The thing on his desk is a wafer holding 10,000 lenses. He likes to point out that across his entire doctorate he hand-built maybe 100 devices, one at a time, in a cleanroom. Then a single foundry shot produced 10,000 in one pass. "In my entire PhD, I made probably 100 different devices. But in one single shot, we made 10,000." That gap - between a scientist's craft and a semiconductor line - is the whole company.

Metalenz makes meta-optics: flat surfaces covered in nanostructures that steer light without any curvature. The payoff is not just thinness. A metasurface can do jobs an ordinary lens cannot - collapse a stack of components into one, capture depth, read polarization, see in near-infrared - while being manufactured on the same lines that turn out the chips in a phone.

Devlin co-founded the company in 2016 with his Harvard advisor Federico Capasso and battery-industry veteran Bart Riley. The trigger was a paper. That year his group showed, for the first time, high-quality images shot through metasurfaces across the visible spectrum. It landed on the cover of Science and was named one of the journal's ten breakthroughs of the year. Investors from the world's largest cellphone makers noticed. The lab decided to spin out.

From 2016 to 2021 the work was unglamorous: finding product-market fit, then convincing a semiconductor foundry that a lens could be treated like a transistor. The chasm between a stunning demo and a shipping product is where most deep-tech companies fall in. Devlin calls it exactly that - "crossing the chasm" - the move from compelling technology to market adoption and scale.

They crossed it in 2022. Metalenz announced the world's first metasurfaces inside consumer devices, in partnership with STMicroelectronics. Roughly five years from a lab bench to mass production. In a field where "flat optics" had been an academic curiosity for two decades, that timeline is the achievement.

Then came Polar ID. Most face-unlock systems flood your face with thousands of infrared dots and read the pattern back. Devlin's team built a sensor that reads the polarization of light reflecting off a face - a signature a photograph or a mask struggles to fake - in a package small and cheap enough for a mid-range phone. It is the first polarization sensor aimed at consumer markets, and it is the kind of capability that only becomes affordable when the optics are printed, not ground.

A basement, a radio, a grandfather

Before the wafers there was a basement. Devlin spent childhood hours there with his grandfather - a World War II radio operator and self-taught engineer - assembling shortwave radios. The moment they switched one on and heard voices arriving from the other side of the planet stuck with him. Technical knowledge, he learned early, is only interesting when it connects to a human being.

His father ran the other half of that lesson. Without a college degree, he climbed to global sales manager at Flexitallic, a gasket manufacturer. Watching him, Devlin absorbed how a salesperson, an engineer, and a machinist all pull toward one commercial goal. It is why he insists that research is a form of selling: "Even as a researcher, you are always selling your research. Ultimately you are communicating your idea to colleagues so that they realize why it's important."

The academic path ran from a joint BS/MS in electrical engineering and materials science at Drexel to a PhD in applied physics at Harvard, in Capasso's group - one of the founding labs of the entire metasurface field. Twenty-plus publications, fifteen-plus patents and applications, and a habit of asking for help: "Don't be afraid to ask questions. There are a lot of people who have done this already and are willing to help."

Billions of devices, no curves

The ambition is plain and large: put meta-optics into billions of devices, retiring the bulky stacks of curved glass that sit behind every camera. Depth sensing for robots and cars. Polarization for security. Near-infrared imaging for health and biometrics. All of it riding the cost curve of the semiconductor industry rather than the cost curve of precision glass.

In early 2025 Optica named him the recipient of its Kevin P. Thompson Optical Design Innovator Award, citing his foundational metasurface design work, his leadership in commercialization, and that first consumer polarization sensor. The same spring, the Harvard Gazette ran a piece on the company under the headline "From Harvard lab to your living room" - which is, more or less, the entire job description.

When he steps away, it is to hike with his family or ride long-distance cycling routes - the kind of thing that lets a founder hold perspective while betting a company on light you cannot see. The wafer stays on the desk. It is not a trophy. It is the moment his idea stopped being only his.