compiler-optimization jobs in San Diego

Design, implement, and verify machine learning software for embedded systems running on ARM processors, DSPs, and GPUs in mobile, edge, automotive, and IoT products. Develop optimized ML kernels, compilers, and tools; optimize ML frameworks like TensorFlow and PyTorch for efficient hardware execution.…See more Perform integration, testing, and performance optimization of embedded software subsystems. Collaborate with hardware teams on joint design and development, coordinate cross-team dependencies, and contribute to design reviews.

AMD seeks a Principal Compiler Architect to lead the design and optimization of LLVM-based compilers for AMD GPUs, focusing on performance improvements for ML frameworks and HPC applications. The role involves working with architecture specialists on hardware bring-ups, debugging compiler issues, and developing groundbreaking compilation technologies for current and future GPU hardware.…See more The candidate will apply data-driven optimization strategies, stay current with hardware and software trends, and collaborate across teams to enhance compiler support for ML and HPC workloads. Required expertise includes strong C/C++ programming, understanding of GPU execution models, and ideally a track record of open-source LLVM contributions.

As a Senior Systems Engineer for Data Center AI at Qualcomm, you will research, develop, optimize, and validate AI/ML solutions that integrate Qualcomm's hardware accelerators with software and ecosystem to deliver high-performance inference in datacenter environments. You will design and deploy Gen AI and LLM applications, implement fine-tuning and distillation techniques, perform model benchmarking, and drive system-level architecture decisions.…See more You will collaborate across functional teams to meet system-level requirements while staying current with advances in AI/ML models and hardware. Required skills include strong Python proficiency, expertise in ML frameworks (PyTorch, TensorFlow), deep understanding of ML deployment and system performance profiling, and experience with parallel computing.