Software Engineer, Inference - CUDA / Kernels

About the Team We’re building high-performance infrastructure to serve OpenAI’s frontier models at massive scale. As part of the inference team, you’ll be responsible for unlocking every last FLOP from our GPUs by designing kernels, tuning memory layouts, and optimizing model execution at the lowest levels of the stack. About the Role We are looking for a kernel-focused engineer to lead efforts in writing, porting, and optimizing GPU kernels used in inference workloads. This role requires deep familiarity with CUDA or equivalent kernel programming environments, and a strong intuition for performance tuning across modern GPU architectures. In this role, you will: Design, implement, and optimize CUDA kernels for inference-critical operations (e.g., fused matmuls, custom activation functions, memory layout transforms). Analyze performance bottlenecks and optimize kernel execution for throughput and latency. Contribute to and extend internal GPU libraries and runtime tools. Work closely with hardware-specific profiling tools (e.g., Nsight, nvprof) to guide improvements. Collaborate with researchers to port or re-architect new model operations for production use. You might thrive in this role if you: Have deep expertise in CUDA, and have written high-performance GPU code used in production systems. Understand GPU memory hierarchies, warp scheduling, and kernel-level tuning. Have experience debugging low-level perf issues with Nsight, cupti, or ROCm tools. Are excited to build optimized compute primitives that scale across fleets of accelerators. Enjoy working closely with model researchers to bring new operators to life. Nice to have: Contributions to GPU kernel libraries or frameworks (Triton, cuBLAS, cutlass). Experience with mixed precision or tensor core optimization. Familiarity with MIG, multi-instance GPU configurations, or NUMA-aware execution. About OpenAI OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of the capabilities of AI systems and seek to safely deploy them to the world through our products. AI is an extremely powerful tool that must be created with safety and human needs at its core, and to achieve our mission, we must encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity. We are an equal opportunity employer, and we do not discriminate on the basis of race, religion, color, national origin, sex, sexual orientation, age, veteran status, disability, genetic information, or other applicable legally protected characteristic. For additional information, please see OpenAI’s Affirmative Action and Equal Employment Opportunity Policy Statement . Qualified applicants with arrest or conviction records will be considered for employment in accordance with applicable law, including the San Francisco Fair Chance Ordinance, the Los Angeles County Fair Chance Ordinance for Employers, and the California Fair Chance Act. For unincorporated Los Angeles County workers: we reasonably believe that criminal history may have a direct, adverse and negative relationship with the following job duties, potentially resulting in the withdrawal of a conditional offer of employment: protect computer hardware entrusted to you from theft, loss or damage; return all computer hardware in your possession (including the data contained therein) upon termination of employment or end of assignment; and maintain the confidentiality of proprietary, confidential, and non-public information. In addition, job duties require access to secure and protected information technology systems and related data security obligations. We are committed to providing reasonable accommodations to applicants with disabilities, and requests can be made via this link . OpenAI Global Applicant Privacy Policy At OpenAI, we believe artificial intelligence has the potential to help people solve immense global challenges, and we want the upside of AI to be widely shared. Join us in shaping the future of technology.