GPU Software Engineer (CUDA)
- Design and implement high-performance CUDA kernels for compute-intensive workloads across AI and HPC use cases.
- Profile and optimize GPU code using tools such as Nsight Systems, Nsight Compute, and CUDA profilers.
- Tune memory access patterns, occupancy, register usage, and shared memory utilization for peak performance.
- Develop highly optimized libraries for linear algebra, attention, and other ML primitives.
- Optimize multi-GPU and multi-node training using NCCL, RDMA, and high-performance networking.
- Implement custom operators and fused kernels in PyTorch, JAX, or Triton.
- Collaborate with ML engineers to identify performance bottlenecks in training and inference pipelines.
- Develop benchmarks and regression tests to safeguard performance over time.
- Evaluate new GPU architectures and feature sets, and advise on adoption strategy.
- Contribute to compiler-level optimizations for tensor programs where appropriate, working at the boundary between ML frameworks and underlying accelerator codegen to unlock performance not reachable through framework-level tuning alone.
- Optimize memory hierarchy usage across HBM, L2, shared memory, and registers.
- Implement mixed-precision and quantized compute paths that maximize accelerator throughput while preserving numerical fidelity within bounds acceptable for the target workloads.
- Document performance characteristics, design decisions, and tuning playbooks for internal teams.
- Stay current with GPU architecture, CUDA evolution, and emerging accelerator technologies.
- Bachelor’s or Master’s degree in Computer Science, Computer Engineering, or a related field.
- Six or more years of experience in GPU programming and performance engineering.
- Deep expertise in CUDA C/C++ and GPU programming models.
- Strong understanding of modern GPU architectures, memory hierarchies, and execution models.
- Hands-on experience profiling and optimizing GPU workloads in production.
- Familiarity with NCCL, MPI, and high-performance interconnect technologies.
- Experience integrating custom kernels into ML frameworks.
- Strong C++ skills and familiarity with modern systems programming practices.
- Solid grounding in linear algebra and numerical methods.
- Strong communication and collaboration skills with research and engineering teams.
- Experience with Triton, CUTLASS, or other GPU kernel authoring frameworks.
- Familiarity with TensorRT, FasterTransformer, or vLLM internals.
- Exposure to compiler infrastructure such as LLVM or MLIR.
- Open-source contributions to GPU or ML performance libraries.
- Experience with large-scale distributed training infrastructure.
Equal Employment Opportunity (EEO) Statement
Bright Vision Technologies (BV Teck) is committed to equal