Performance Engineer, GPU

About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems. About the role: Pioneering the next generation of AI requires breakthrough innovations in GPU performance and systems engineering. As a GPU Performance Engineer, you'll architect and implement the foundational systems that power Claude and push the frontiers of what's possible with large language models. You'll be responsible for maximizing GPU utilization and performance at unprecedented scale, developing cutting-edge optimizations that directly enable new model capabilities and dramatically improve inference efficiency. Working at the intersection of hardware and software, you'll implement state-of-the-art techniques from custom kernel development to distributed system architectures. Your work will span the entire stack—from low-level tensor core optimizations to orchestrating thousands of GPUs in perfect synchronization. Strong candidates will have a track record of delivering transformative GPU performance improvements in production ML systems and will be excited to shape the future of AI infrastructure alongside world-class researchers and engineers. You might be a good fit if you: - Have deep experience with GPU programming and optimization at scale - Are impact-driven, passionate about delivering measurable performance breakthroughs - Can navigate complex systems from hardware interfaces to high-level ML frameworks - Enjoy collaborative problem-solving and pair programming - Want to work on state-of-the-art language models with real-world impact - Care about the societal impacts of your work - Thrive in ambiguous environments where you define the path forward Strong candidates may also have experience with: - GPU Kernel Development: CUDA, Triton, CUTLASS, Flash Attention, tensor core optimization - ML Compilers & Frameworks: PyTorch/JAX internals, torch.compile, XLA, custom operators - Performance Engineering: Kernel fusion, memory bandwidth optimization, profiling with Nsight - Distributed Systems: NCCL, NVLink, collective communication, model parallelism - Low-Precision: INT8/FP8 quantization, mixed-precision techniques - Production Systems: Large-scale training infrastructure, fault tolerance, cluster orchestration Representative projects: - Co-design attention mechanisms and algorithms for next-generation hardware architectures - Develop custom kernels for emerging quantization formats and mixed-precision techniques - Design distributed communication strategies for multi-node GPU clusters - Optimize end-to-end training and inference pipelines for frontier language models - Build performance modeling frameworks to predict and optimize GPU utilization - Implement kernel fusion strategies to minimize memory bandwidth bottlenecks - Create resilient systems for planet-scale distributed training infrastructure - Profile and eliminate performance bottlenecks in production serving infrastructure - Partner with hardware vendors to influence future accelerator capabilities and software stacks Deadline to apply: None. Applications will be reviewed on a rolling basis. The expected salary range for this position is: The annual compensation range for this role is listed below. For sales roles, the range provided is the role’s On Target Earnings ("OTE") range, meaning that the range includes both the sales commissions/sales bonuses target and annual base salary for the role. Annual Salary: $280,000—$850,000 USD Logistics Minimum education: Bachelor’s degree or an equivalent combination of education, training, and/or experience Required field of study: A field relevant to the role as demonstrated through coursework, training, or professional experience Minimum years of experience: Years of experience required will correlate with the internal job level requirements for the position Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices. Visa sponsorship: We do sponsor visas! However, we aren't able to successfully sponsor visas for every role and every candidate. But if we make you an offer, we will make every reasonable effort to get you a visa, and we retain an immigration lawyer to help with this. We encourage you to apply even if you do not believe you meet every single qualification. Not all strong candidates will meet every single qualification as listed. Research shows that people who identify as being from underrepresented groups are more prone to experiencing imposter syndrome and doubting the strength of their candidacy, so we urge you not to exclude yours

Strong candidates will have a track record of delivering transformative GPU performance improvements in production ML systems and will be excited to shape the future of AI infrastructure alongside world-class researchers and engineers. You might be a good fit if you: - Have deep experience with GPU programming and optimization at scale - Are impact-driven, passionate about delivering measurable performance breakthroughs - Can navigate complex systems from hardware interfaces to high-level ML frameworks - Enjoy collaborative problem-solving and pair programming - Want to work on state-of-the-art language models with real-world impact - Care about the societal impacts of your work - Thrive in ambiguous environments where you define the path forward. Strong candidates may also have experience with: - GPU Kernel Development: CUDA, Triton, CUTLASS, Flash Attention, tensor core optimization - ML Compilers & Frameworks: PyTorch/JAX internals, torch.compile, XLA, custom operators - Performance Engineering: Kernel fusion, memory bandwidth optimization, profiling with Nsight - Distributed Systems: NCCL, NVLink, collective communication, model parallelism - Low-Precision: INT8/FP8 quantization, mixed-precision techniques - Production Systems: Large-scale training infrastructure, fault tolerance, cluster orchestration

Architect and implement foundational GPU performance systems for Claude; maximize GPU utilization and inference efficiency at large scale; develop cutting-edge optimizations from low-level tensor core code to distributed multi-node orchestration; work across the stack from hardware interfaces to high-level ML frameworks; collaborate with researchers and engineers to deliver transformative performance improvements; contribute to design of distributed communication strategies and resilient, planet-scale distributed training infrastructure