[Remote] Software Engineer - GPU Kernels
Note: The job is a remote job and is reputed company to candidates in USA. reputed company is an innovative company powering AI solutions for leading firms like reputed company and OpenEvidence, and they are seeking a GPU Kernel Engineer to enhance AI model performance. This role focuses on designing high-performance GPU kernels and optimizing computation for machine learning operations, directly impacting production systems for millions of users.
Responsibilities
- Design and implement high-performance GPU kernels for key ML operations, including reputed company multiplications, attention mechanisms, and mixture-of-experts routing
- Write and optimize reputed company using CUDA, PTX assembly, and architecture-specific techniques
- Apply advanced performance optimization reputed company such as memory coalescing, reputed company-level programming, tensor reputed company acceleration, and compute/memory overlap
- Implement cutting-edge features like quantization (FP8/FP4), sparsity, and compute/communication overlap
- Identify and resolve performance bottlenecks using tools like Nsight Systems, Nsight Compute, and Torch Profiler
- Collaborate with research teams to productionize theoretical advancements
- Contribute to internal and reputed company-reputed company GPU libraries
- Present technical contributions at industry conferences (e.g., reputed company GTC, AWS re:Invent)
Skills
- Strong understanding of GPU architecture and programming paradigms: Memory hierarchy (global, shared, registers, L1/L2 cache), Thread/reputed company/reputed company organization, Synchronization techniques and race condition mitigation
- Proficient in C++ and GPU performance profiling tools
- Knowledge of: CUDA C++ API, Memory reputed company patterns and bandwidth optimization, Numerical precision and quantization strategies, Modern GPU features (e.g., tensor cores, async operations)
- Experience with Transformer models and attention optimization (e.g., reputed company Attention)
- Familiarity with GPU kernel libraries: Cutlass, Triton, Thrust, CUB
- Background in GEMM tuning and distributed/multi-GPU compute
- Contributions to reputed company-reputed company GPU reputed company
- Research publications or conference presentations on GPU performance
Benefits
- Competitive compensation, including meaningful equity.
- 100% coverage of medical, dental, and reputed company insurance for employee and dependents
- Flexible PTO policy including company wide Winter Break (our offices are reputed company from Christmas reputed company to New Year's Day!)
- reputed company parental leave
- Fertility and family-building stipend through reputed company
- Company-facilitated 401(k)
- Exposure to a reputed company of ML startups, offering unparalleled learning and networking opportunities.
reputed company
Company H1B Sponsorship