Job Description:
As a staff ML engineer, you will work within a team, build and adapt ML models for Edge devices to be used as blueprints to facilitate the building of customer products. You will be a pioneer in model execution and testing on high performance computing subsystems and help craft the engineering experience in the development of data driven applications on Edge and Automotive devices.
Responsibilities:
- Your day-to-day role will involve experimentation, research, model development, model deployment and performance monitoring for machine learning applications at the Edge.
- Define the appropriate model architecture and fine-tune the model for an optimum application performance.
- Collaborate with global multi-functional teams, including hardware engineers, firmware developers, product managers, tools teams and system architects, to deliver integrated solutions.
- Occasional travel e.g., visits to global Arm offices and developer conferences.
Required Skills and Experience:
- Consistent track record in development of machine learning and deep learning applications; solid knowledge in popular Machine learning frameworks, including PyTorch, TensorFlow and scikit-learn.
- Excellent programming skills preferably in R, Python, Jupyter, C/C++ and Matlab. Able to adapt and add new functionality to sophisticated models.
- Proficiency with technical methods for Hardware-aware model optimization and compression including quantization, pruning and knowledge distillation.
- Education: Degree or equivalent experience in a relevant subject, such as mathematics, data science or computer science.
“Nice To Have” Skills and Experience:
- Experience with graphics, vision and machine learning APIs and technologies.
- Expertise in usage of ML frameworks and runtimes for the Edge including TensorFlowLite / TFLite Interpreter, PyTorch / ExecuTorch and ONNX / ORT
- Hands-on development of Large Language models, vision transformer models and Generative AI applications; knowledge in Retrieval Augmented Generation, fine-tuning and prompt engineering optimization methods.
- Familiarity with ARM architecture and ARM based AI hardware accelerators.