Senior Software Engineer, CUDA Deep Learning Systems

This is a fully remote position, open to applicants in California, +1 more state.

📋 Description

• Investigate, analyze, and prototype system enhancements for cutting-edge deep learning models that bridge high-level deep learning frameworks and low-level CUDA through modeling, simulation, and silicon prototyping.

• Design and optimize distributed computing architectures, ranging from single-node setups to cluster-scale supercomputing environments.

• Create, implement, and enhance custom high-performance CUDA kernels tailored for new neural network architectures and workloads.

• Examine hardware-software interactions to pinpoint and address performance bottlenecks in training and inference workflows.

• Collaborate with AI researchers, hardware and software architects, kernel and compiler developers, and CUDA driver specialists to co-design systems and algorithms.

• Develop exploratory tools and runtime systems aimed at profiling and accelerating innovative deep learning paradigms.

• Produce clean, efficient, and maintainable code, transitioning prototypes into open-source releases, framework integrations, internal tools, or commercial products.

⛳️ Requirements

• BS, MS, or PhD in Computer Science, Computer Engineering, Electrical Engineering, or a related discipline, or equivalent experience.

• 8+ years of pertinent industry experience or comparable academic experience following degree completion.

• Strong expertise in C++ and Python programming languages.

• Solid understanding of deep learning principles, particularly focusing on transformers.

• Comprehensive knowledge of distributed computing, multi-node scaling, and performance challenges at the cluster scale.

• Demonstrated experience in systems programming, computer architecture, and low-level systems performance optimization.

• Familiarity with GPU accelerator architectures.

• Practical experience with CUDA programming, kernel optimization, and workload profiling.

• Experience in profiling and optimizing generative AI models, including large language models.

• Research experience in machine learning systems or related fields.

• Experience in profiling and optimizing vision models, generative AI architectures, or diffusion models.

• Proven track record of initiative and eagerness to tackle problems across the technology stack.

• Preferred: expertise in the performance internals and execution graphs of PyTorch, JAX, TensorRT, vLLM, sgLang, Nemo, or Megatron.

• Preferred: experience with NCCL, MPI, UCX, and distributed machine learning methodologies such as pipeline, tensor, or expert parallelism.

• Preferred: knowledge of numerical methods and low-precision arithmetic, including NVFP4, MXFP4, FP8, or INT8.

• Preferred: background in deep learning compilers and ML systems, such as Triton, XLA, or torch.compile.

• Preferred: experience in designing agentic AI systems for complex systems and infrastructure challenges.

🏝️ Benefits

• Equity

• Benefits

• Equal opportunity employer

• Inclusive work environment