Embedded deep learning for a better night’s sleep? This might be the coziest-sounding use of a Qualcomm Snapdragon processor we’ve ever heard of.
Meet Qualcomm Developer of the month Evan Shapiro, and his company Knit Health, Inc.
Knit Health was born from a group of new parents, interested in new tools and technology to help parents in the consumer space. Evan says ‘We (co-founders Evan, Paul and David) worked at global design firm IDEO for over 10 years. We have experience in product design and engineering for many consumer and healthcare domains. We decided to combine our shared expertise in science, engineering, and product design to launch Knit Health.’
We caught up with Evan to talk product design and poker…
Tell us a little about your company.
Knit Health is building a smart camera to measure health. Our first product is designed for parents of young children, and measures sleep and respiration. The camera is based on the Snapdragon 410 processor, with embedded deep learning model and signal processing. The launch client app for beta users is iOS only, with Android version coming next year. We currently have a Kickstarter campaign to sell our first units to early adopters.
What advice would you give to other developers?
Don’t spend too much time designing and wireframing the ideal solution. Focus on tools and methods that get you something, anything, that works as fast as possible. Your conception of what should be done changes so much after taking the first steps, that a lot of the effort you put in to architecting the ideal long-term solution is wasted. When building something new you should be in the mindset of rapid prototyping, failing early, and iterating quickly.
What makes your company culture unique?
Five of the nine full-time employees are parents of young children. This is unique for San Francisco startups, and we have a family friendly culture. You don’t have to be a group of 22-year-olds willing to put in 14 hour days at the office to build something amazing.
Share with us a fun fact about yourself?
I used to be a professional internet poker player, which was the inspiration for my first company, which built tools and subscription data services for professional internet poker players.
What do you love about embedded and IoT development?
Being able to work across the full stack, from metal to cloud to client, we have end-to-end control over everything that makes up our end-user experience.
Where do you and your team get inspiration for your work?
From games, to dev blogs, to science publications, we all share a genuine passion for science and engineering.
When enduring a long day, how do you and your team stay energized?
Go to the coffee shop for a hot chocolate, the dumpling shop for a bite, or the brewpub for a beer- San Francisco is a great place to work!
Where do you see the IoT industry in 10 years?
“IoT” is a temporary term, that only makes sense in the context of a world filled with ‘dumb’ things. In 10 years, the ‘Internet of Things’ industry will just be the ‘Things’ industry.
What Qualcomm technologies are featured in your projects?
Our smart camera is built on the Snapdragon processor, and runs Linux (Linaro).
How does the Snapdragon processor assist in the final product?
It’s the backbone. The 410 is a standout in its value class, and provides the power we need at a cost that is suitable for a consumer grade product.
Did using this specific Qualcomm technology speed up your development process?
The DragonBoard 410c and 96Boards (Linaro) greatly aided in speeding our BSP, kernel, and driver level development.
Follow @knithealth on Twitter or visit the Knit Health website for more information.