Key Responsibilities :

- Model conversion pipelines (e.g., PyTorch ONNX)

- Inference runtimes : ONNX Runtime, TensorRT, ROCm/HIP

- Toolchains and compilers like TVM, XLA, Vitis AI, XDNA

- Kernel execution, memory transfer, and result validation

- Profile ML workloads and optimize for performance.

- Work on validation plans across software lifecycle stages.

- Collaborate with cross-functional and global teams on end-to-end AI/ML pipeline validation.

Required Skills :

- Strong hands-on experience with Linux (Ubuntu) and Yocto.

- Solid understanding of machine learning concepts : deep learning, LLMs, recommender systems.

- Proficiency with Python programming.

- Experience in ML frameworks : PyTorch, TensorFlow, ONNX Runtime.

- Experience profiling AI/ML workloads.

- Prior exposure to GPU/Accelerator validation stacks (HIP, CUDA, OpenCL, OpenVINO).

- Familiarity with production-level QA methodologies and defect lifecycle management.

- Excellent analytical and problem-solving skills.

- Strong communication skills and sense of ownership.