Member of Technical Staff, ML Compiler and Systems (Bay Area, Remote)

What You'll DoLead the evolution of our high-performance robotics simulation platformDesign and implement the compute infrastructure and data flow mechanisms to optimize performance for physics simulation and foundation model trainingLead development of our compiler stack, focusing on JIT compilation, LLVM IR, and GPU codegen to minimize compile time and maximize runtime performanceCollaborate with the team to improve the compiler's support for differentiable programming, crucial for training neural networks within simulationsStay current on state-of-the-art ML compilers—for example those in torch, Triton, and JAX—and decide which techniques and approaches are best suited for our applicationWork closely with simulation and robotics engineers to align compiler enhancements with application needsContribute to relevant open-source projects and participate actively in the broader compiler and systems communityWhat You'll BringStrong background in compiler construction, particularly in JIT compilation and LLVM-based code generationExtensive experience with GPU programming models (e.g., CUDA, Vulkan) and an understanding of GPU architectureTrack record as a core contributor to GPU programming infrastructure—such as Torch, JAX, Mojo, Taichi, or WarpProven ability to profile and optimize complex systems for performance and scalabilityUnderstanding of automatic differentiation and its application in simulation and machine learning contextsExcellent communication skills and a collaborative approach to problem-solvingEnthusiasm for contributing to and engaging with open-source communities#J-18808-Ljbffr