AI Accelerator Software Principal Engineer - Inference

Ampere; Portland (OR); On-site; USD 182,000 - 273,000; Full time

Benefits offered by this job

Health insurance

401(k) retirement plan

Unlimited flextime

10+ paid holidays

Job summary

Ampere is seeking an AI Accelerator Software Principal Engineer – NPU Full‑Stack Integration to lead high‑performance, low‑latency inference on Arm Ethos U85. You will optimize software stacks from model execution to accelerator kernels and collaborate with hardware teams for scalable AI workloads.

The role demands strong C/C++, Python, and Linux systems expertise, plus ML understanding and modern AI tooling fluency. Expect a competitive base pay and comprehensive benefits.

Qualifications

• BS in CS/CE/EE or related field with 8+ years experience; or MS 6+ years; or PhD 3+ years.
• Experience with AOT compilation in PyTorch and edge deployment paths.
• Linux accelerator/runtime development experience preferred.

Responsibilities

• End-to-end deep learning performance acceleration.
• Model enablement with quality and speed for edge devices.
• Hardware/software co-design and optimization with platform teams.
• Build AI software components and optimized execution paths.
• Cross-functional collaboration with compiler, kernels, and product teams.

Skills

Python

C/C++

Performance profiling

Linux development

AI/ML concepts

Education

Bachelor's degree in CS/CE/EE

Master's degree

PhD

Tools

PyTorch

Compiler/runtime

Linux kernel

Job description

Ampere is seeking an AI Accelerator Software Principal Engineer – NPU Full‑Stack Integration to lead high‑performance, low‑latency inference on Arm Ethos U85. You will optimize software stacks from model execution to accelerator kernels and collaborate with hardware teams for scalable AI workloads.

The role demands strong C/C++, Python, and Linux systems expertise, plus ML understanding and modern AI tooling fluency. Expect a competitive base pay and comprehensive benefits.