Principal RTL Design Engineer, IO & Memory Systems

SambaNova is seeking a Principal Engineer to own the design, integration, and implementation of high-speed IO and memory subsystems within the DataScale® platform. This role is central to delivering reliable, high-performance AI systems at scale.

You will drive these subsystems across the full product lifecycle, from architecture exploration and microarchitecture definition through RTL and integration of internal and third-party IP (PCIe, Ethernet , CXL, HBM, die-to-die). This is a hands-on role for an engineer with deep protocol and RTL expertise, who can operate in ambiguity and execute with high ownership., * Engage in architecture exploration with system and chip architects to define IO and memory subsystem direction

• Drive feasibility studies and design trade-off analysis to meet PPA constraints
• Own microarchitecture definition and specification for IO and memory subsystems
• Deliver synthesizable, high-quality SystemVerilog RTL for assigned blocks against functionality, performance, and power targets
• Run the front-end design flow end to end, covering lint, CDC/RDC, synthesis, timing constraint development, and design-for-test readiness
• Partner with STA and physical design teams on timing closure, floorplanning, congestion analysis, and optimization at advanced process nodes
• Work closely with verification and emulation teams to ensure design validation coverage and quality
• Collaborate with software and firmware teams to ensure optimal end-to-end solutions
• Drive cross-functional triage of complex functional and performance issues to root cause and closure
• Apply engineering best practices including code review, testing, and refactoring
• Improve design and integration methodology, reuse, and flow scalability

• MS in Electrical Engineering, Computer Science, or a related field
• 7+ years of experience in ASIC design
• Experience in logic design, microarchitecture, specification, SystemVerilog RTL design, and functional verification
• Experience with power, performance, or area optimization
• Experience with two or more on-chip or chip-to-chip protocols and interfaces such as AXI, APB, PCIe, Ethernet, or UCIe
• Experience with memory subsystems such as HBM and CXL
• Ability to evaluate and select vendor IP, and to read and distill complex IP and protocol specifications into actionable design requirements
• Clear and effective communicator with a proactive, collaborative approach to problem-solving, * Experience taking one or more ASIC products from concept to silicon
• Familiarity with emerging scale-up protocols (ESUN, SUE, UAL) and scale-out protocols such as Ultra Ethernet
• Experience designing on-chip interconnect and traffic-management logic, including NoC fabrics, switches, arbiters, and quality-of-service or congestion control mechanisms
• Knowledge of machine learning and hardware accelerator architectures

SambaNova offers a competitive total rewards package, including the base salary, plus equity and benefits. We cover 95% premium coverage for employee medical insurance, and 77% premium coverage for dependents and offer a Health Savings Account (HSA) with employer contribution. We also offer Dental, Vision, Short/Long term Disability, Basic Life, Voluntary Life, and AD&D insurance plans in addition to Flexible Spending Account (FSA) options like Health Care, Limited Purpose, and Dependent Care. Our library of well-being benefits available to you and your dependents includes a full subscription to Headspace, Gympass+ membership with access to physical gyms, One Medical membership, counseling services with an Employee Assistance Program, and much more.

SambaNova is a leader in next-generation AI infrastructure, delivering a full-stack inference platform for customers worldwide. At the core of SambaNova’s technology is the RDU (Reconfigurable Dataflow Unit) - a chip built on a dataflow architecture rather than the traditional GPU model. Its decode performance is especially strong for agentic workloads like multi-turn agents, code generation, and long-running applications. RDUs are packaged into SambaRack, rack-scale hardware that lets customers deploy state-of-the-art models with better performance, greater energy efficiency, and faster time to value.