GPU Optimization Engineer - San Francisco, CA; Oakland, CA

DoorDash Labs is an independent team within DoorDash. We explore robotics and automation to transform last mile logistics in the long term. If you have a passion for applying robotics solutions in a service used by millions of people, then we want to talk to you!

You will own high performance kernel libraries for GPUs. There are kernels related to neural networks and kernels for optimizing C++ code.

You will report to a Senior Engineering Manager on our Autonomy team in our Labs organization. We are open to someone working remotely, or hybrid in our offices in the San Francisco Bay area.

• Design, develop and optimize GPU kernels for latency critical code. 
• Use your knowledge of the microarchitecture of the GPU to map C++ algorithms to high performance GPU kernels.
• Identify opportunities for optimization in a sophisticated autonomous robot stack.
• Immediately see the impact of your work on the robot.
• Have the opportunity to learn about the motion planning stack.

• Experience with core kernels such as matrix multiply.
• Industry experience implementing kernels for GPUs directly in CUDA.
• Experience with mapping non uniform workloads to the GPU.
• Deep experience with the GPU toolchain: compiler, profilers and debuggers.
• Strong C++ and CUDA ability 

At DoorDash, our mission to empower local economies shapes how our team members move quickly, learn, and reiterate in order to make impactful decisions that display empathy for our range of users—from Dashers to merchant partners to consumers. We are a technology and logistics company that started with door-to-door delivery, and we are looking for team members who …