Software Development Engineer - AI Workload Optimization - TX, Austin, United States
Originally posted 28 October 2026 by the employer.
About the role
As a Software Development Engineer in the AI Software group, you will define and drive the end-to-end software optimization strategy across the AMD software stack—including ROCm, compilers, and high-level AI frameworks—to achieve maximum performance for AI workloads.
What you'll do
- Set the technical vision and roadmap for workload optimization across the AI software stack. - Lead the profiling, analysis, and tuning of large-scale models, including LLMs, Diffusion, Multimodal, and MoE architectures, on AMD hardware. - Partner with customers and hyperscalers to evaluate workload requirements and deliver architectural enhancements and software optimizations. - Collaborate across hardware architecture, compiler, and framework teams to inform future silicon features based on AI workload trends. - Drive the development of tools and frameworks for performance estimation, modeling, and automated reporting. - Act as a technical ambassador in industry forums and open-source communities while mentoring engineers.
What you'll need
- Experience in software development with technical leadership in a high-level role (Fellow or equivalent). - Bachelor's degree or Master's degree in Computer Science, Electrical Engineering, or a related field, or equivalent experience. - Experience in AI hardware architecture, software optimization, and mapping model architectures to low-level software. - Experience solving performance bottlenecks and leading cross-functional engineering initiatives.
Nice to have
- Expertise in AI Frameworks including PyTorch, JAX, vLLM, SGLang, and the ROCm software stack. - History of optimizing distributed inference and training at scale across multi-node and multi-GPU environments. - Experience with performance profiling tools, including TorchProfiler, ROCm Profiler, or Nsight, and hardware-level performance modeling. - Understanding of model architectures (Transformer, Attention, KV Cache) and optimization techniques including quantization, speculative decoding, and FlashAttention. - Research or applied experience in AI/ML, such as deep learning, model training/inference optimization, large language models, or computer vision.
Skills: ROCm, Workload Optimization, LLMs, PyTorch, Distributed Training
This role has been open 0 days — well below the 56-day median for Infrastructure/Platform Software roles.
Infrastructure/Platform Software · GPU Software Stack
|
Open roles in category
247
|
Median days open
56 d
|
Median salary
$221k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | AMD | All employers we track in this specialty (247 roles · 20 employers) |
|---|---|---|
| Open roles in this specialty | 51 | 247 |
| Open roles in the wider Software, Firmware & Systems family | 385 | 6389 · 147 employers |
| Median days open | 57 d | 56 d (+1 d vs this employer) |
| Median salary (USD postings) | — | $221k |
Skills observed across this category: ROCm, Workload Optimization, LLMs, PyTorch, Distributed Training
Who's hiring in this category
- NVIDIA · 126 open roles · median 57 d
- AMD (this employer) · 49 open roles · median 56 d
- Intel Corporation · 27 open roles · median 15 d
- Qualcomm · 17 open roles · median 187 d
- Arm Holdings · 5 open roles · median 125 d
- Graphcore · 4 open roles · median 7 d
How we counted: 247 open Infrastructure/Platform Software (GPU Software Stack) roles from 20 employers tracked in the SemiconductorJobs index, counted 28 Oct 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there — it is counted in the wider Software, Firmware & Systems family row. Figures refresh nightly.