Principal Software Engineer (Runtime)

Working at SambaNovaThis role presents a unique opportunity to shape and work on cutting-edge system software solutions for AI and machine learning applications in the enterprise & commercial landscape. The stack spans multiple software layers, and provides products & services including but not limited to OS, software-hardware interface, isolation through virtualization and containerization, scaleout through data-parallel/model-parallel/pipeline-parallel, etc with networking, fault management, application orchestration using Kubernetes, and various tools for management/debugging/diagnostics.Job DescriptionSambaNova is searching for a runtime software engineer candidate that will deliver cutting-edge system software solutions for AI and machine learning applications in the enterprise & commercial landscape. The candidate will participate in building, testing, and deploying next-generation high-performance OS-related system software stack for compute systems for AI applications at scale. The candidate must have a strong background in programming, building, and testing software in distributed systems, networking, and system fault management with good teamwork and planning skills.This individual will be responsible for:Design, develop and maintain OS-related system software with industry-leading performance and capabilities.Work across teams to ensure maximum system-level performance.Innovate new software-hardware interfaces to manage compute, memory, network and I/O resources for AI applications.Own and support testing automation of the RDU system software of different configurations and use casesProvide feedback to the hardware teams on potential improvements Basic QualificationsBachelor's or Master's Degree in Computer Science, Computer Engineering6-12 years of related industry, hands-on experienceExperience in programming languages, such as C, C++, Python Additional Required QualificationsExperience and knowledge with I/O virtualization technologies, industry-standard open-source networking, and parallel computing frameworks such as MPI, and UCX is a bonus.Experience with Linux kernel, drivers, user space libraries, and applications with strong Linux fundamentalsExperience with PCIe, DMA is a plus.Demonstrated knowledge of Ethernet, DMA, RDMA, RoCE concepts. Knowledge of system fault management is desiredGood communication skills and enthusiasm to help colleaguesKnowledge of various compiler and build systems, such as GCC, LLVM, Make, and CMake Preferred QualificationsExperience working in a high-growth startupA team player who demonstrates humilityAction-oriented with a focus on speed & resultsAbility to thrive in a no-boundaries culture & make an impact on innovation PRB