Tenstorrent
Datacenter & Agentic AI Workload Performance Analysis Engineer
$100K–$500K
Santa Clara, California, United States · Full-time · Posted Oct 1, 2026
- Experience
- No specific number of years stated; welcomes candidates at various experience levels.
- Education
- PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related field.
- Visa sponsorship
- Employment is contingent on eligibility to access U.S. export-controlled technology; citizenship/permanent residency status or prior license approval may be required, and an offer may be rescinded if employment is not possible under U.S. export laws.
About the job
Analyze datacenter and agentic AI workloads on RISC-V platforms, identify CPU performance opportunities, and inform architecture through workload characterization, modeling, and hardware-software collaboration.
What you'll do
- Bring real-world workloads to RISC-V platforms, characterize their behavior, and identify performance, efficiency, and scalability bottlenecks.
- Reduce production workloads into representative workloads and traces for performance modeling and architectural exploration.
- Correlate simulation results with hardware behavior and translate findings into recommendations for CPU architects, RTL designers, software engineers, and compiler teams.
What they ask for
- PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related field, with strong research or industry experience in workload characterization, benchmark development, performance analysis, or simulation.
- Deep understanding of CPU architecture and RISC-V, including pipelines, speculative execution, vector/SIMD extensions, and memory hierarchies.
- Hands-on experience with performance analysis and simulation tools such as Linux perf, strace, QEMU, or CPU microarchitecture simulators.
- Strong programming skills in C/C++, Python, Bash/Shell, and assembly or intrinsic programming; strong understanding of systems software and GNU/RISC-V software ecosystems.
Pay
$100k - $500k (including base and variable compensation targets; not identified as base salary or total pay)
Tools
RISC-V, Linux perf, strace, QEMU, C/C++, Python, Bash/Shell, GNU/RISC-V
AI focus
Analyze datacenter and agentic AI workloads to guide RISC-V CPU architecture and improve performance, scalability, and efficiency.
Model-written summary · the original posting is authoritative