Tenstorrent
Software Engineer, Acceleration Kernel Development
$100K–$500K
Santa Clara, California, United States +2 · Full-time · Posted Jul 23, 2024
- Location
- Santa Clara, California, United States
Toronto, Ontario, Canada
Toronto, Ontario, Canada - Experience
- Candidates at various experience levels are welcome; candidates are assessed for an appropriate level during interviews.
- Visa sponsorship
- Employment is contingent on eligibility to access U.S. export-controlled technology. Depending on citizenship/permanent residency status, prior license approval from the U.S. Commerce Department or applicable federal agency may be required.
About the job
Develop and optimize low-level acceleration kernels that improve performance for machine-learning workloads on AI hardware.
What you'll do
- Build and optimize compute kernels for parallel ML and high-performance workloads.
- Analyze and tune instruction-level performance across latency, memory, and bandwidth.
- Debug, profile, and maintain low-level software, integrating optimizations with ML engineers.
What they ask for
- Expertise building and optimizing compute kernels for parallel ML and high-performance workloads.
- Ability to analyze and tune instruction-level performance across latency, memory, and bandwidth.
- C/C++ proficiency to build fast, efficient code.
- Collaborative approach to integrating optimizations into production.
Pay
$100k - $500k (base and variable compensation targets; combined range, not stated as base salary or total pay)
Tools
C/C++
AI focus
Develop and tune kernels for machine-learning workloads and integrate kernel work into ML frameworks and training pipelines.
Model-written summary · the original posting is authoritative