Staff Tech Lead Manager, Deep Learning Compiler

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 heavily relies on machine learning techniques to perform a variety of tasks, each with different requirements of hardware/compute resources. Throughout the life-cycle of each machine learning model, skilled ML engineers (on both training and inference sides) work closely to prepare it for a robust, scalable, and compute/power efficient inferencing on a resource-constrained hardware accelerator. Such a close working relationship is key to fast and successful deployment of intelligent systems on the car.

Cruise is looking for a deep learning compiler engineer to build the compiler and software tool chain for deploying machine learning models on to a variety of ML hardware accelerators. In this position, you will contribute, develop and enhance Cruise’s internal ML compiler infrastructure for high-performance and retargetability possibly by leveraging open-source technology like LLVM, TVM and XLA.

In this role, you will collaborate closely with engineers from different AV Engineering teams (e.g. Computer Vision, Perception, platform) to scope out system/software requirements at the application level while engaging with AV hardware teams to understand the target hardware platform and its constraints.

If you're interested in optimizing machine learning inference on different hardware accelerators, and want to test your skills with real-world (and practical) applications in the autonomous vehicle domain, let's chat!

Day-to-day responsibilities include: 

• Design, implement and test compiler features and capabilities related to IR infrastructure and compiler passes
• Develop graph compiler optimizations like operator fusion, layout optimization, etc that are customized to each of the different ML accelerators in the system
• Integrate open-source and vendor compiler technology in to Cruise’s internal compiler infrastructure
• Build performance tooling to evaluate, understand and improve ML performance on different accelerators
• Collaborate with cross functional agile teams of AV engineers to guide the direction of inferencing and provide requirements and feature requests for hardware vendors
• Closely follow industry and academic developments in the ML compiler domain and provide performance guidelines and best practices for other ML engineers

You should apply for this role if you have the following qualifications:

• At least one of the following
• Strong research record in compiler domain and 3+ years experience in compiler design
• 5+ years experience with retargeting LLVM, or equivalent retargetable compiler infrastructure.
• Significant contributions to an open source compiler frameworks like TensorFlow, TVM, PyTorch etc.
• 2+ years of experience with deep learning
• Experience with deep learning frameworks (e.g., Tensorflow, etc) and software stack (e.g., TensorRT, TVM, XLA etc)
• Experience with ML accelerators and hardware architecture
• Strong expertise in writing production quality C++ code
• Comfortable and experienced in software development lifecycle - coding, debugging, optimization, testing, integration
• Familiarity with parallelization techniques for ML acceleration
• MS, or higher degree, in CS/CE/EE, or equivalent, in industry experience

Bonus points!

• GPU programming (CUDA) and familiarity with deep learning stack (e.g., cuDNN, cuBLAS)
• SIMD programming (avx2, neon)

The salary range for this position is $193,900 - $285,000. Compensation will vary depending on location, job-related knowledge, skills, and experience. You may also be offered a bonus, restricted stock units, and benefits. These ranges are subject to ch