Principal Place and Route (PnR) Engineer

You will develop and tune the partitioning, placement, and routing algorithms that map customer workloads onto the RDU. You will work with the compiler team, hardware architects, and performance engineers to model new hardware constraints and trace throughput gaps back to mapping decisions. Your changes will ship in production releases and show up in measured performance.

The Place and Route team writes the software that maps dataflow graphs onto SambaNova’s Reconfigurable Dataflow Unit (RDU). Our tools partition each graph, place the pieces on the chip’s compute and memory units, and route the communication between them. The quality of that mapping sets the throughput of the compiled program, including large language model inference and training. We are looking to grow our team with highly motivated engineers who want to utilize advanced algorithms to solve challenging problems.

• Develop partitioning, placement, and routing algorithms that meet hardware resource constraints and keep compile times practical
• Tune PnR results for target workloads, including LLM inference at different batch sizes and concurrency levels
• Model new hardware constraints and architecture features in our solvers and heuristics as the chip evolves
• Analyze compiled output with performance engineers and trace throughput gaps back to mapping decisions
• Build the infrastructure for prototyping, regression testing, and deployment of PnR changes
• Work with the compiler and hardware teams to define interfaces and resource models

• M.S. or Ph.D. in Computer Science, Computer Engineering, or related field
• 3+ years of experience developing optimization or graph algorithms in industry or research
• Expert knowledge of graph algorithms, partitioning algorithms, combinatorial optimization, or MILP solvers
• Strong proficiency in C/C++
• Familiarity with Python
• Experience designing large, highly parallel, performance-sensitive systems
• Ability to learn hardware architecture and apply it to improve performance
• Experience solving NP-hard problems in practice with heuristics, approximation, or pruned exact methods

• Experience with place and route for FPGAs, CGRAs, or other spatial architectures
• Experience with ML compilers such as MLIR, LLVM, TVM, or XLA
• Experience with simulated annealing, constraint programming, or SAT/SMT solvers
• Familiarity with LLM inference stacks such as vLLM and PyTorch
• Publications at venues such as FPGA, DAC, ICCAD, or MLSys

Base Pay Range

$1—$999,999 USD

Submission Guidelines Please note that in order to be considered an applicant for any position at SambaNova Systems, you must submit an application form for each position for which you believe you are qualified.

Benefits Summary for US-Based, Full-Time Employment Positions

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.