Principal Deep Learning Comm. Architect

Job Description

Job Title: Principal Deep Learning Communication Architect. Location: Santa Clara, CA, USA; Austin, TX, USA; Remote (USA)

Responsibilities:

• Define the long-term technical roadmap for communication libraries across next-generation platforms.
• Enable seamless scaling of AI models to clusters with hundreds of thousands of nodes.
• Lead the design and development of next-generation communication primitives and collective algorithms.
• Optimize communication for heterogeneous interconnects such as NVLink, Ethernet, and InfiniBand;
• Collaborate with application developers to design specialized communication primitives for AI and HPC workloads.
• Ensure AI and HPC libraries such as NCCL, NVSHMEM, UCC, and UCX evolve to support large-scale models and advanced AI systems.
• Work closely with hardware and software teams to influence next-generation networking hardware specifications.
• Develop analytical models and simulators to predict system performance under emerging workloads.

Qualifications:

• PhD or MS in Computer Science, Electrical Engineering, or a related field, or equivalent experience.
• 12 or more years of industry experience in high-performance computing or distributed deep learning.
• Deep understanding of parallelism techniques including data, tensor, and pipeline parallelism;
• Expertise in advanced parallelism strategies such as context parallelism, expert parallelism, and ZeRO optimizations.
• Strong technical proficiency with NCCL, UCX, UCC, NVSHMEM, or MPI.
• Experience with RDMA, RoCE, and InfiniBand technologies;
• Advanced knowledge of inference engines such as TensorRT-LLM, vLLM, or similar systems.
• Strong understanding of GPU architecture, including memory hierarchy and CUDA programming models.

Preferred:

• Experience developing frameworks such as Megatron-Core, DeepSpeed, or JAX/XLA.
• Contributions to open-source projects such as PyTorch Distributed, KServe, or Ray;
• Proven experience deploying and optimizing large-scale models on GPU-based systems.
• Publications or patents in top-tier systems or architecture conferences.

Benefits and Compensation:

• Base salary range: $272,000 to $431,250 per year, depending on location and experience.
• Eligibility for equity and comprehensive benefits package;
• Opportunity to work with cutting-edge AI and HPC technologies in a high-impact environment.