Product at Google; fmr founder/CEO of Bump
|Recommendations||1 person has recommended David|
Join LinkedIn & access David's full profile
I'm a product- and engineering-focused entrepreneur who likes to solve real problems for normal people with simple products.
My focus: Make the world a better place by building products people want.
Some of my thoughts on mobile product design:
Follow me on Twitter at https://twitter.com/dflieb
Subscribe to my updates at http://www.facebook.com/david.lieb
Google acquired Bump in September 2013.
Origami was acquired by eFamily in November 2013. The Origami team was acquired by Nest, and subsequently Nest was acquired by Google.
Origami (formerly Everyme) is a YC-funded private social networking startup. Origami has created two products: Everyme, a private mobile social networking app; and Origami, a private social network for families.
Bump was acquired by Google in September 2013.
Helping more than 150 million people interact in the real world:
Share photos, files, contacts, and more between mobile devices and between phones and computers by just bumping.
Flock knows which family and friends you are with when photos are taken and brings everyone's photos into a single shared album. Flock is one of the first "inference-based" apps that require zero user intent post-install.
Our design philosophy:
Bump was funded by Y Combinator, SV Angel (Ron Conway), Sequoia, and Andreessen-Horowitz (Marc Andreessen).
Explored new opportunities for TI's MEMS technologies in a cross-disciplinary skunkworks team of 10. Built prototypes for new products including:
- 3D imaging via structured light
- Motion-adaptive handheld projectors
- Heads-up displays
I helped make DLP display systems brighter, sharper, more colorful, and thinner. Matlab was my best friend. I built prototypes out of cardboard over Christmas break, filed more than a dozen patent applications, and killed it on the TI softball team.
My major projects included:
- algorithms for solid state illumination systems
- laser TVs
- cell phone projectors
Full-time at TI from June 2005 through August 2008 following 4 summer internships in various groups at TI, including two summers in the Kilby Fab advanced R&D center.
I helped make robots see while working in Sebastian Thrun's Artificial Intelligence Lab. I also did a stint in Marc Levoy's Graphics Lab building algorithms for indoor mapping. My focus was novel *practical* applications of computer vision and machine learning techniques to make real things happen for real people (and robots).
We call this "Duty Cycle On-The-Fly". Make projectors and TVs brighter by dynamically adjusting the colors and timing in real time based on scene content. So if your scene is mostly red, don't waste time and energy displaying lots of blue and green light.
We call this "Off State Light Recycling". When a pixel in the image you are displaying is dark, you are wasting all the light incident on that pixel. Here's how you recycle that light back into the bright pixels and make your display brighter.
A robot drives through a forest. It eventually reaches terrain it can't navigate (a tree, a swamp). Local sensors tell it that (ex: it hits a tree). Then it looks backwards in time to see what that tree looked like when it was far far away. Then it teaches itself to remember what that tree looked like and avoid it in the future. This is what nerds call Self-supervised Learning; our twist is to use optical flow techniques to go backwards in time.
A. Lookingbill, J. Rogers, J. Curry, D. Lieb, and S. Thrun. Reverse optical flow for self-supervised adaptive autonomous robot navigation. International Journal on Computer Vision (IJCV), 2007.
A helicopter flies above a scene on the ground. It's moving all over the place, so is the stuff on the ground. Using really smart algorithms, the helicopter can figure out what is moving on the ground using just a simple video camera. Then, after watching the ground for a little while, it can learn what types of things typically happen on the ground below it -- where people walk, when to expect busses, whether something out of the ordinary is occurring.
A. Lookingbill, D. Lieb, D. Stavens, and S. Thrun. Learning activity-based ground models from a moving helicopter platform. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Barcelona, Spain, 2005.
You and your friends love to take photos together,
but they always end up on different phones.
Flock finds the photos that you've taken together
and makes it easy for you to share them with each
other. So there's no more forgetting to share photos.
Producer, Director, Writer, Editor
Shot on RED Epic by Dale Tibor
Music by Mira Music
VFX by Aaron Bartlett http://www.linkedin.com/pub/aaron-bartlett/a/44/a18
Color Grading by Creative Rebellion
Over 300 million professionals are already on LinkedIn. Find who you know.
Product Manager, Android Camera at Google
Engineering Manager at Google
Engineer at Google
Director of Operations and Support at Humble Bundle
Business Strategy and Operations at Tango Me
Product Manager at Google
Product, Marketing, and Growth
ceo at Bump Technologies (acquired by Google)
VP Product at Google
Associate Vice President at Penn State University
State College, Pennsylvania Area
Greater Boston Area
Account Executive at Ally Financial Inc.
Kansas City, Missouri Area
Laboratory Assistant at Touch Tek Laboratories
Washington D.C. Metro Area
Independent Motion Pictures and Film Professional
London, United Kingdom