Staff GPU Engineer, Compute Kernels [Remote Friendly]

Key Skills

Aartificial intelligence, Analytical and Problem solving, API, C++, CUDA kernels, CUDA/GPU programming, Effective communication skills, Machine learning techniques, Mentoring, OpenCL, Optimization

Job Description

We're Cruise, a self-driving service designed for the cities we love.

We’re building the world’s most advanced self-driving vehicles to safely connect people to the places, things, and experiences they care about. We believe self-driving vehicles will help save lives, reshape cities, give back time in transit, and restore freedom of movement for many.

In our cars, you’re free to be yourself. It’s the same here at Cruise. We’re creating a culture that values the experiences and contributions of all of the unique individuals who collectively make up Cruise, so that every employee can do their best work. 

Cruise is committed to building a diverse, equitable, and inclusive environment, both in our workplace and in our products. If you are looking to play a part in making a positive impact in the world by advancing the revolutionary work of self-driving cars, come join us. Even if you might not meet every requirement, we strongly encourage you to apply. You might just be the right candidate for us.

About the role: 

The Autonomous Vehicle (AV) software stack is a sophisticated workload that includes compute and data intensive functions within perception, prediction, planning and control. This role is within the ML Accelerators (MLA) team which is responsible for maintaining and optimizing the software stack for real-time, on-road performance. This means efficient workload mapping onto a sophisticated, resource-constrained hardware, co-architecting the HW platform with HW engineers and building platforms and tools for AI deployment and performance profiling, diagnosis and optimizations that can be employed by AV engineers (from perception, prediction, planning and controls). MLA engineers work in close collaboration with AV engineers throughout the lifecycle of crafting new AV features which is essential to the success of our mission.

About the team:

MLA's Kernels team is looking for a Staff GPU Engineer to scale our technology stacks and compute kernels. In this position, you will be responsible for creating the strategic vision and owning the design and development of high-performance and reliable GPU-accelerated libraries of primitives for Cruise’s current and next generation custom silicons. Your solutions will be used by Compiler or Domain-Specific libraries to implement the core AV functions across ML and non-ML workloads spanning multiple GPU architectures. As Cruise expands our vehicle offerings on our path to productization, your libraries will not only enable Cruise’s low cost custom silicon critical to our long term business needs, but also provide the needed performance portability to ensure our successful scaling.

If you're interested in this role, let's chat!

What you’ll be doing:

• Build a long-term strategic roadmap to enable Cruise to migrate custom GPU-accelerated codes to libraries of primitives for OpenCL backend
• Architect, design and develop GPU libraries and APIs that are optimized across multiple GPU architectures (custom accelerator inside AV and GPGPUs in cloud).
• Collaborate closely with our Compiler and Runtime teams to build scalable APIs for core primitive compute libraries
• Deliver production-quality, reusable and reliable high-performance compute libraries to be used by cross-functional teams (Compiler, Performance Tooling, Runtime, ML Framework and AV Development)
• Own the performance of GPU-accelerated kernels and libraries (both on and off vehicle), as AV SW stack evolves
• Review code submissions, provide guidance on design and implementation, break down tasks for junior developers, track execution and efficiently report status.
• Create and implement technology standards, methodologies, processes, and guidelines
• Coach and mentor team members on engineering excellence projects.

What you must have:

• Excellent GPU programming skills in OpenCL with a thorough understanding of the OpenCL programming model and GPGPU architecture
• Strong background in software architecture, library and API design, and design patterns.
• Strong C++ and object-oriented programming skills, with experience in developing in large codebases.
• Hands-on experience building (and optimizing) accelerated libraries of compute primitives (e.g. DNN, BLAS, GEMM)
• Hands-on experience benchmarking, profiling and optimizing GPU-accelerated libraries and kernels
• Solid background in system performance, high performance computing and architecture-aware optimizations.
• Strong communication, influence, and negotiation skills and the ability to work collaboratively within a team
• Exceptional leadership skills and ability to set direction and work through ambiguities
• Excellent analytical and problem-solving skills
• Minimum 5+ years of relevant industry or equivalent experience