Staff Machine Learning Engineer
About the Role
The Central Technology ML (CTML) team pioneers and implements the technology roadmap for the future of machine learning computing on Arm architecture. This role involves joint development across hardware, software, and ML algorithms to balance constraints and drive innovation. As a member of the CTML Content and Algorithm team, you will contribute to the Center of Excellence for Machine Learning by analyzing complex neural networks on Arm’s computing platforms and developing algorithms to advance industry leadership.
Responsibilities
- Develop an in-depth understanding of current and future ML workloads on Arm compute platforms, focusing on PPA (performance, power, and area).
- Serve as the ML Algorithm expert in architecture exploration, collaborating on joint hardware, software, and algorithm development.
- Drive algorithm innovation and prototyping to validate architectural features.
- Create hardware-aware optimization algorithms to enable state-of-the-art networks on Arm platforms, including CPU, GPU, and NPUs.
Requirements
- 3+ years of experience in ML Algorithm Development and/or Optimization.
- A Master’s or PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related technical field.
- Proficiency in computer architecture with basic knowledge of hardware and software design.
- Experience developing and working with large software systems in languages like Python.
- Knowledge of deep learning libraries such as TensorFlow and PyTorch.
- Experience training large deep learning models on GPU-based systems.
Preferred Skills
- ML Model Optimization techniques for resource-constrained Arm edge compute platforms.
Benefits
Arm offers a competitive total reward package, including salary and additional benefits. The company fosters a collaborative and innovative environment with a 10x mindset, encouraging high performance and personal well-being. Hybrid working arrangements are available, tailored to team needs.
Pay
Salary range: $249,900–$338,100 per year.