Research Engineer - AI Performance & Kernel Optimization

Zyphra is an artificial intelligence company based in San Francisco, California.The Role:As a Research Engineer - AI Performance & Kernel Optimization, you will improve and optimize the performance of our large-scale language model training and inference stacks. You will work closely with our pretraining and inference teams to identify bottlenecks, design and implement highly optimized kernels, and push the limits of throughput, latency, and hardware utilization across a range of accelerator platforms. This role is suited for someone who enjoys deep systems work, cares about performance at every level of the stack, and is excited to translate low-level optimizations into meaningful gains for frontier-scale AI systems.You’ll Work Across:Kernel development and optimization for large-scale ML workloads, using any level of the stack from PTX/assembly to CUDA, HIP, Triton, or other GPU DSLsPerformance tuning for training and inference stacks across GPUs and other acceleratorsProfiling and eliminating bottlenecks in memory movement, communication, scheduling, and compute utilizationOptimizing distributed training and inference systems for large MoE models, including large-scale model parallelismPortability and optimization across non-NVIDIA hardware, with special interest in AMD hardware such as the MI300x and MI355xCollaboration with research and infrastructure teams to turn systems improvements into real-world model training and inference gainsWhat We're Looking For / Requirements:Strong engineering aptitude for building reliable, high-performance systemsExcellent low-level performance intuition and the ability to reason about hardware-software interactionsAre excited to rapidly learn new systems, tools, and hardware environmentsExcellent communication and collaboration skills, with the ability to work effectively across research and engineering teamsEnjoy diving deep into the weeds and hunting down the last 10–20% of performanceQualifications / Additional Skills:Experience writing highly performant GPU kernels at any level of abstraction–PTX, CUDA, HIP, Triton, or other kernel DSLsExperience optimizing ML workloads for large-scale training, ideally in language model pretraining or inference environmentsExperience with non-NVIDIA accelerator hardware, such as AMD, AWS Trainium, Google TPU, Qualcomm, ARM, Intel, and custom ASICsStrong understanding of distributed training systems and parallelism schemes, including data parallelism, tensor/model parallelism, pipeline parallelism, sharding, and communication/computation overlapExperience with performance engineering in other demanding parallel computing environments such as HPC, quantitative finance, scientific computing, graphics, compilers, or numerical simulationStrong systems intuition around memory hierarchy, bandwidth constraints, kernel fusion, launch overhead, communication overhead, and hardware utilizationExperience using profiling and debugging tools to drive performance improvementsFamiliarity with infrastructure underlying large-scale training and inference, including collective communication libraries, and runtime performance analysisBackground in a highly technical field such as physics, mathematics, theoretical computer science, computer science, or electrical engineeringAny HPC experience is a strong plusWhy Work at Zyphra:Our research methodology is grounded in methodical, step-by-step approaches to ambitious goals. Both deep research and engineering excellence are equally valuedWe strongly value new and crazy ideas and are very willing to bet big on new ideasWe move as quickly as we can; we aim to minimize the bar to impact as low as possibleWe all enjoy what we do and love discussing AIBenefits and Perks:Comprehensive medical, dental, vision, and FSA plansCompetitive compensation and 401(k) planRelocation and immigration support on a case-by-case basisIn-office snacks and meals providedUnlimited PTO and company holidaysIn-person team in San Francisco with a collaborative, high-energy environment