Arm
Principal AI Performance Engineer
$263K–$355K
San Jose, California · Principal · Full-time · Posted Sep 23, 2026
- Visa sponsorship
- The role does not meet eligibility requirements for sponsorship; the successful candidate must have the right to work in the US without relying on sponsorship by Arm.
About the job
Optimize production AI workloads and DNN inference on Arm technology for customer applications, improving performance and power efficiency on edge devices.
What you'll do
- Develop kernel- to system-level optimized solutions and production-quality reference implementations, documentation, and technical content.
- Work with customers and internal engineering teams to diagnose and resolve performance issues.
- Use customer insights to influence Arm’s IP and software roadmaps.
What they ask for
- Experience optimizing DNNs in Triton, CUDA, or another kernel-level programming language.
- Deep understanding of parallel computing, memory hierarchies, and DNN performance optimization.
- Strong Python and C++ skills; experience with modern AI frameworks, execution models, and profiling and analysis tools.
- Strong communication and interpersonal skills.
Nice to have
- Customer-facing or field engineering experience; experience in the Arm ecosystem.
- Background in AI performance optimization for edge devices.
Pay
USD $262,700-$355,400 per year (base salary)
Tools
Triton, CUDA, Python, C++
AI focus
Optimize DNN kernels and production AI inference workloads for performance and power efficiency on Arm technology, including edge devices.
Model-written summary · the original posting is authoritative