Description

You will develop and integrate AI software components across ROCm, RyzenAI, and embedded AI runtimes.

Responsibilities

• Contribute to operator enablement, runtime features, and framework integrations for PyTorch, ONNX Runtime, TensorFlow.
• Implement and debug GPU/NPU execution paths, including hybrid CPU+GPU+NPU workflows.
• Support integration of AI models and pipelines on RyzenAI APUs, ROCm GPUs, and AMD Instinct accelerators.
• Enable and validate AI models across CV, NLP, LLM, and multimodal domains.
• Perform functional validation, accuracy analysis, and performance benchmarking (latency, throughput, memory).

Required Skills

• 8+ years of experience in software development, AI/ML engineering, or system validation.
• Strong programming skills in Python and C/C++.
• Solid understanding of Linux, build systems, and debugging tools.
• Hands-on experience with AI frameworks such as PyTorch, TensorFlow, or ONNX Runtime.
• Experience running and debugging AI models on GPU-accelerated systems.
• Experience integrating AI software components across ROCm and RyzenAI.
• Familiarity with heterogeneous computing (CPU+GPU+NPU).

Key Skills

Education

Any Gradute