Join our GPU Assembler team to craft cutting-edge tools for high-performance applications in HPC, DL, and Autonomous Driving. You will collaborate with global HPC developers, enhancing their GPU performance and efficiency on current and future NVIDIA GPUs through industry-leading Assembler and Code Analysis Tools. This role is integral to developing a core component for GPU-accelerated general-purpose computation. Our Disassembler (NVDISASM) is a vital tool for CUDA users worldwide, offering extensive code analysis. The Assembler is crucial for optimizing key DL applications.

Lead the design and development of NVIDIA's GPU Assembler and Disassembler. Implement binary analysis and instrumentation features, including call graph generation, register usage tracking, and GPU binary patching. Work closely with GPU architecture and debugger/profiler teams to deliver new features and product enhancements. Collaborate with related component teams to ensure seamless integration, reliability, and high-quality code generation. Engage with customers and partners to gather feedback, driving innovative ideas and features into the product.

A BS or MS degree in Computer Science, Computer Engineering, or a related field is required, along with 8+ years in low-level system software development. A minimum of 6 years of experience in compilers, assemblers, binary utilities, and binary analysis tools is essential. Strong analytical and C/C++ programming skills are necessary. Proficiency in at least one area of compiler development, including feature support, code generation, or compiler infrastructure, is expected. A solid understanding of Assembly Language, Processor ISA (GPU ISA is a plus), and processor architecture is also required. Familiarity with object file formats like ELF and debugging formats such as DWARF is needed.