By the time he was 15, Catalin Voss produced the No. 1 podcast on the Germany iTunes stores and commuted back and forth between his native Heidelberg and Silicon Valley to work for Steve Capps, one of the designers of the original Macintosh Apple computer.
During freshman year at Stanford University, the coding prodigy added “CEO” to his résumé.
His company, Sension, is a visual interface company seeking to revolutionise the way we learn. Voss, 19, and roughly eight employees (mostly Stanford graduate students) developed a facial recognition software that works with a simple webcam, allowing anyone to make videos that respond to the viewer.
In the early days of Sension’s development, Voss realised that all his peers were bringing laptops, tablets, and other devices equipped with monocular webcams to class. “Why can’t you interact with them in a more sophisticated way?” Voss asked himself. So he set out to solve a problem in online education: boredom.
If you’re watching a lecture on your laptop, Sension’s software might prompt you with a question if it senses you’re not paying attention, or explain a subject matter further if you appear confused. The product can verify the identity of the test-taker, making cheating nearly impossible, and provide teachers with unparalleled analytics on what questions stumped students the most.
Voss built out the business in the education sector and gained several customers, including Mindflash, an online business training software company. Another unnamed customer will help roll out Sension’s technology in all the major standardized testing associations and universities in the next year.
“That is Sension, the business,” Voss says, the money-maker. But that first summer after entering Stanford, Voss experienced what he calls his “ah-ha” moment.
“We had built a lightweight face-tracker, something that could track and understand many points in a person’s face. We could use it on a mobile device, so why shouldn’t it run on what’s essentially a cheap Android phone tied to your head?”
He imported the software into a Google Glass app. It worked, and was able to identify the feelings of the subject that the wearer looked at. Suddenly the uses for his product doubled, and he realised that the software had vast applications in the autism community.
People on the autistic spectrum often have difficulty in judging the emotional expression of faces. While little is known about how individuals with autism misinterpret this information, we know the impairment worsens over time, according to new research from Georgetown University.
The “ah-ha” moment hit home for Voss, whose cousin was diagnosed with autism and has trouble recognising facial expressions. He was familiar with the educational tools used at the time, namely flashcards, and decided Sension products could provide those social cues in real time.
While wearing the computerized eyewear, the user locks in on a person’s face. The display then reads the person’s emotion and reports it to the user by audio or text. “Happy.” “Surprised.” “Upset.”
In summer 2013, Sension was admitted to the prestigious [email protected] program, a collegiate startup accelerator sponsored by Highland Capital Partners. An $US18,000 grant allowed Voss and his cofounder, Stanford sophomore Jonathan Yan, to hire coders and continue development on these two applications of Sensions: one for online education and one for the autism community.
Voss says he then reached a crossroads. “We had built all this interesting stuff, and had to ask, what are we going to do now? People asked, ‘Are you going to drop out of school?'” Voss says, adding that Sension received numerous acquisition offers. He resolved to stay at Stanford and split his time between being a student and running a company at the age of 19.
The university ecosystem offered him support he wasn’t ready to walk away from. Stanford provides work space, advisers, funding (the School of Medicine donated a “couple hundred thousand dollars” toward research and development), and most importantly, the ability to conduct clinical trials.
This month, Sension began clinical studies with 40 participants. “We’re trying to figure out how to build something that they actually want to use,” Voss says. They’re asking questions like, does visual or audio feedback make more sense? Should the app alert the user of a person’s expression or wait until they have made eye contact to make the identification? He hopes kids from the clinical trial will take the devices home by the end of the year to test them in their everyday environments.
In the meantime, Voss will continue to grow both divisions of Sension and go where his passion takes him. He will be one to watch in Silicon Valley, considering he’s caught the eyes of Wired, The Dish Daily, USA Today, SF Gate, and more media outlets before being old enough to buy a drink at a bar.
He remains humble. When I asked Voss for his best hiring advice, as an underclassman interviewing doctorate candidates for his startup, his response surprised me.
“I look for people who are better at stuff and smarter than me. I believe that’s a good practice in general,” Voss laughs. “It works well when you’re young — everyone is older than you and has more experience.”