Senior Principal Runtime Engineer - San Jose, California, United States

Originally posted 24 March 2026 by the employer.

About the role

The Runtime Team builds a high-performance, distributed, and scalable software execution environment for SambaNova SambaStack and Cloud platforms to support data-flow applications such as ML training, inference, and HPC applications. This role focuses on developing runtime stacks, drivers, and user-space libraries supporting AI, ML, and scientific workloads across distributed compute systems.

What you'll do

- Build and enhance runtime stack features to support high-performance, scalable ML inference and training applications. - Develop drivers and kernels interfacing directly with hardware and custom silicon accelerators. - Eliminate networking bottlenecks across distributed systems using switching, routing, and congestion management techniques. - Develop user-space libraries to maximize utilization and performance of hardware resources. - Create user-facing tools for Datascale systems, including analysis, job management, hardware management, profiling, and debugging utilities. - Collaborate cross-functionally across Hardware, ML Application, Compiler, and DevOps teams.

What you'll need

- B.S. in Computer Science, Engineering, or a related field. - 5+ years of software engineering experience emphasizing distributed systems, networking, or cloud infrastructure. - In-depth knowledge of runtime stacks, user-space libraries, and tuning software for high-performance distributed systems. - Solid understanding of switching and routing, with the ability to configure and debug switches and routers. - Hands-on experience with RDMA and RoCE networking stacks, including RDMA-based verbs and congestion management. - Hands-on experience developing kernel drivers and system-level software interfacing directly with hardware, device registers, and device memory management. - Experience with software development and CI tooling including Git, Jenkins, and Jira.

Nice to have

- Experience designing or working closely with custom hardware accelerators (ASICs, FPGAs) and understanding low-level interactions. - Knowledge of SDN, DPDK, SONiC, or distributed communication libraries (UCX, MPI, NCCL). - Familiarity with deploying high-performance systems in distributed, cloud, or data center environments.

Skills: AI and machine learning applications, system software, Operating System interface, Networking, distributed systems, scalable software