All jobsApply now
AI Accelerator Software Principal Engineer - Framework Integration
Ampere Computing
Portland, OR · RemoteFull-time$182,000 – $273,000 / yearly
Job details
Pay
$182,000 – $273,000 / yearly
Job type
Full-time
Work setting
Remote / Hybrid
Location
Portland, OR
Posted
Posted 2 months ago
Full job description
Hybrid role in Portland, OR. Lead design and delivery of high-performance deep learning inference solutions for Ampere's AI accelerator.
What you will do:
- Integrate and optimize frameworks like PyTorch, ONNX, and llama.cpp for accelerator backends
- Optimize across the full stack: serving, framework layers, compilers, runtimes, and kernels
- Ensure model compatibility with production serving ecosystems including vLLM and SGLang
- Collaborate on hardware/software co-design to improve throughput, latency, and resource efficiency
- Develop reusable libraries and integrate AI capabilities into cloud-native platforms
- Work cross-functionally with compiler, kernel, platform, and product engineering teams
Qualifications
- User-mode driver or GPU/accelerator runtime development experience on Linux
- Knowledge of Transformer and Diffusion model architectures
- Fluency with AI development tools like Codex or Claude Code
Requirements
- BS in Computer Science/Engineering or related field with 8+ years experience (or MS with 6+ years, or PhD with 3+ years)
- Strong experience with PyTorch, llama.cpp, and ONNX framework integration
- Expert-level proficiency in Python and C/C++
- Demonstrated performance profiling and tuning expertise
- Solid AI/ML fundamentals including neural networks
SemiconductorAI
Ready to apply?
Apply on the company siteYou’ll be taken to an external site to complete your application.