Tenstorrent

Datacenter & Agentic AI Workload Performance Optimization Engineer

$100K–$500K

Santa Clara, California, United States +2 · Full-time · Posted Oct 5, 2026

Apply on Tenstorrent's site Save on Drimva

Location: Santa Clara, California, United States
Toronto, Ontario, Canada
Santa Clara, California, United States
Education: Master’s or 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; depending on citizenship/permanent residency status, prior license approval may be required.

About the job

Optimize datacenter and agentic AI workloads on next-generation RISC-V platforms through runtime, compiler, software, and hardware-aware performance engineering.

What you'll do

Bring up and tune runtimes; profile applications and investigate performance bottlenecks in memory and concurrency.
Develop and evaluate optimizations using RISC-V Vector/Matrix capabilities and custom instructions.
Build AI-assisted and automated workflows for profiling, coding, benchmarking, and optimization.

What they ask for

Master’s or PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related field, with strong experience in performance optimization, computer architecture, compilers, or systems software.
Hands-on runtime or compiler optimization experience, such as OpenJDK/JIT, LLVM, GCC, V8, Python, or equivalent systems.
Strong understanding of CPU performance, memory hierarchies, concurrency, garbage collection, vector/SIMD optimization, and RISC-V architecture.
Expertise with performance profiling and analysis tools; strong programming skills in Java, Python, C/C++, and RISC-V assembly.

Pay

$100k - $500k (total pay, including base and variable compensation targets)

Tools

Java, Python, PHP, Node.js, Lua, Go, Rust, RISC-V

AI focus

Optimize agentic AI workloads and use AI tools to automate profiling, coding, benchmarking, and optimization.

Model-written summary · the original posting is authoritative