Do you dream of tackling a perception challenge with a small team, on an application that is in the real world, making a massive, positive impact on people’s health every day? Are you excited to add perception to a production system where each incremental amount of capability you add directly enables a critical service to save more lives? Do you love looking at the whole problem, including how you are going to validate the solution and find a technical architecture that is tractable to implement? When finding a failure mode, are you eager to go deep into the hardware, data, testing, software pipelines - wherever - to understand it?
At Zipline, there are doctors and millions of patients who rely on us for urgent access to vital medical supplies. Do you want to use perception to enable scaling the delivery of lifesaving value to the billions of people on the planet, starting today? Do you want to work on an application where incremental perception improvements can provide real world value to a live operation now?
As the lead on our Perception team, you will be responsible for developing a perception solution that enables a highly maneuverable aircraft to dodge power lines that are not on any map, avoid buildings that aren’t shown on the satellites photos, stay clear of other aircraft and safely find the right open area to drop life saving supplies.
This will entail finding creative ways to shorten the distance between the team and our customers, guiding the team to make the right trades between classical and machine learning algorithmic approaches, and selecting sensor types in smart ways that work practically for our size, weight, compute, power, and cost constraints. You’ll be building the team over time to ensure that we have all the engineering disciplines required to not only develop the perception system, but to validate it, ship it, and learn from the operational data in order to improve and extend it.
We’re a high-commitment and high-initiative culture, and we’re driven to serve every human being on the planet equally. Today, we fly the equivalent of once around the equator every week. In a few months, we'll be doing that once a day, and by the end of 2020, once every hour.
By owning perception, you will be critical to Zipline’s success in achieving this scale and beyond.
We expect team leads to:
Have experience on a high-performance team in a driven, fast-paced environment where team members have a lot of freedom and autonomy
Be capable of hiring, coaching, and mentoring high-caliber team members, including those better than you in their craft
Work directly with the owners of different parts of the system to integrate perception fully and seamlessly
Be obsessed with shipping and understand what it takes to ship a perception product
Understand how to design and build the simplest, most robust possible solution
Useful Technical Skills / Experience on the bleeding edge of:
While specific technical skills below are useful, we value your ability to align math, algorithms, and code to be able to implement solutions in the real world
Computer vision-based SLAM/multiview geometry
Dense stereo reconstruction
CNNs and deep learning applied to vision problems
Ideally the combination of above 2 (deep learning applied to geometric vision)
Stereo vision, optical flowReal time, multi-threaded/process robotics software systems (c/c++)
We value diversity at Zipline and welcome applications from those who are traditionally underrepresented in tech. If you like the sound of this position, but you are unsure if you meet all the requirements listed above, please apply!
More than two billion people lack adequate access to essential medical products, such as blood and vaccines, due to challenging terrain and gaps in infrastructure.
Zipline improves access to these supplies by flying over impassable mountains and washed-out roads, delivering directly to remote clinics. We centralize supply and provide on-demand deliveries, completely reducing waste and stock-outs.
Zipline provides a seamless delivery system at an affordable price, obsessing over every detail, so you can focus on patient health.