Principal Place and Route (PnR) Engineer
SambaNova · Austin, Texas, United States; San Jose, California, United States · Austin, TX · San Jose, CA
Infrastructureonsiteprincipal$1–$1000k
At a glance
Highlights
- Map dataflow graphs onto SambaNova's RDU to set throughput for LLM inference.
- Tune partitioning, placement and routing for LLM inference at varying batch sizes.
- Ship PnR changes in production releases with directly measured performance impact.
What they offer
- Equity
- 95% premium coverage for employee medical insurance
- 77% premium coverage for dependents
- Health Savings Account (HSA) with employer contribution
- Dental, Vision, Disability, Life and AD&D insurance plus FSAs
- Headspace subscription, Gympass+ membership and One Medical membership
- Employee Assistance Program counseling services
Ideal candidate
Motivated problem-solver with graduate training who loves turning NP-hard mapping challenges into practical, high-throughput solutions and collaborating with compiler and hardware teams.
Requirements
- M.S. or Ph.D. in Computer Science, Computer Engineering, or related field
- 3+ years developing optimization or graph algorithms in industry or research
- Expert knowledge of graph, partitioning, combinatorial optimization or MILP solvers
- Strong proficiency in C/C++ with familiarity in 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 with heuristics, approximation or pruned exact methods
Lab profile
About SambaNova
Full-stack AI infrastructure with RDU dataflow chips for fast agentic inference.
What they build
- RDU Reconfigurable Dataflow Unit
- SambaRack rack-scale hardware
- Full-stack AI inference platform
Known for
- Reconfigurable dataflow architecture alternative to GPUs
- High decode performance for agentic workloads
- Full-stack hardware-software co-design
Key facts
- Founded in 2017
- Headquartered in Palo Alto, California
Posted · Last seen