Research Engineer - AI Performance & Kernel Optimization

Location: On-site in San Francisco, United States

Company

hirify.global develops frontier-scale AI systems and large-scale language model training and inference infrastructure.

What you will do

• Develop and optimize kernels for large-scale machine learning workloads using PTX, assembly, CUDA, HIP, Triton, or other GPU DSLs.
• Tune training and inference stacks across GPUs and other accelerator platforms.
• Profile and eliminate bottlenecks in memory movement, communication, scheduling, compute utilization, and kernel execution.
• Optimize distributed training and inference for large mixture-of-experts models, including model and parallelism strategies.
• Improve portability and performance on non-NVIDIA hardware, including AMD MI300x and MI355x accelerators.
• Collaborate with research and infrastructure teams to translate systems improvements into practical model training and inference gains.

Requirements

• Experience writing highly performant GPU kernels with PTX, CUDA, HIP, Triton, or another kernel DSL.
• Experience optimizing machine learning workloads for large-scale training, language model pretraining, or inference.
• Strong understanding of distributed training and parallelism, including data, tensor/model, pipeline parallelism, sharding, and communication/computation overlap.
• Strong systems intuition covering memory hierarchy, bandwidth constraints, kernel fusion, launch overhead, communication overhead, and hardware utilization.
• Experience with profiling and debugging tools, collective communication libraries, and runtime performance analysis.
• Background in physics, mathematics, theoretical computer science, computer science, electrical engineering, or another highly technical field.

Nice to have

• Experience with AMD, AWS Trainium, Google TPU, Qualcomm, ARM, Intel, or custom ASIC hardware.
• Performance engineering experience in HPC, quantitative finance, scientific computing, graphics, compilers, or numerical simulation.
• HPC experience.

Culture & Benefits

• Research and engineering excellence are valued through methodical, step-by-step work toward ambitious goals.
• New and unconventional ideas are encouraged, with a willingness to invest in high-potential approaches.
• In-person collaboration in a high-energy San Francisco environment.
• Medical, dental, vision, and FSA plans, plus a 401(k) plan.
• Unlimited PTO, company holidays, in-office snacks and meals, and case-by-case relocation and immigration support.