Senior Principal Runtime Engineer
SambaNova Systems8 months ago
Responsibilities
- Develop new and enhanced runtime features for high-performance, scalable ML inference and training applications.
- Build drivers and kernels for next-generation silicon.
- Eliminate networking bottlenecks in high-performance distributed systems.
- Develop user-space libraries that maximize hardware utilization.
- Create user-facing analysis, job-management, hardware-management, profiling, and debugging tools for Datascale systems.
- Collaborate cross-functionally with Hardware, ML Application, Compiler, and DevOps teams.
- Build, test, tune, and deploy high-performance compute systems for AI applications at scale.
Requirements
- Bachelor of Science in Computer Science, Engineering, or a related field.
- At least 5 years of software engineering experience, often focused on distributed systems, networking, or cloud infrastructure.
- Proven experience building, testing, and tuning distributed high-performance systems, including user libraries and runtime stacks.
- Strong understanding of switching and routing, including configuring and debugging switches and routers for application performance.
- Significant experience with RDMA and RoCE networking stacks, RDMA-based verbs, and congestion management.
- Hands-on experience with kernel drivers and system-level software that interfaces directly with hardware.
- Expertise designing and optimizing massively parallel systems for ML training and inference workloads.
- Deep understanding of hardware-software interaction, including registers, device memory management, and accelerator design.
- Familiarity with distributed systems architecture, networking, communication protocols, and scalable compute resources.
- Hands-on experience with Git, Jenkins, and Jira, including automation and continuous integration efforts.
- Ability to design systems that optimize performance and reliability across hardware and software.
- Preferred experience with custom hardware accelerators such as ASICs and FPGAs.
- Preferred knowledge of SDN, DPDK, SONiC, UCX, MPI, NCCL, or modern networking frameworks and distributed communication libraries.
- Preferred experience deploying high-performance systems in distributed, cloud, or data center environments.
Benefits
- US-based full-time employees receive a competitive total rewards package including base salary, equity, and benefits.
- Medical insurance covers 95% of employee premiums and 77% of dependent premiums, with an employer-contributed Health Savings Account.
- Dental, vision, short- and long-term disability, basic life, voluntary life, AD&D, and flexible spending account options are available.
- Well-being benefits include Headspace, Gympass+ access to physical gyms, One Medical membership, counseling services, and an Employee Assistance Program.
About SambaNova Systems
SambaNova Systems builds full-stack AI infrastructure—custom accelerator chips, systems, and foundation models—delivered as on-prem appliances and cloud APIs for enterprises, governments, and data centers. Its offerings include the SambaNova Suite, Samba Cloud, and on-prem SambaStack, powered by its Reconfigurable Dataflow Unit (RDU) architecture, including the SN40L chip. Founded in 2017 and headquartered in San Jose, California, the company is privately held.