Role Overview
The CT-ML Algorithms team drives ML compute on Arm across hardware, software, and algorithms. As a Machine Learning Intern, you will analyze ML workloads, develop algorithms and optimization techniques for performance, power, and area (PPA) on current and future Arm platforms, and build internal tooling for architecture exploration.
Responsibilities
- Develop deep understanding of ML workloads and hardware-aware optimization techniques
- Conduct ML workload analysis to support Arm IP development phases
- Prototype algorithms and optimized models for architecture exploration
- Create internal tooling for algorithmic and architecture exploration
- Automate deployment of HW-aware optimized network processes on Arm processors
Requirements
- Graduate student (PhD or MS) in Computer Engineering, Electrical Engineering, Computer Science or related field
- Deep knowledge of machine learning, deep learning, neural network design, optimization, and compression
- High-level knowledge of computer architecture, systems, and HW-SW co-design
- Proficiency in Python for large software systems
- Experience with TensorFlow and PyTorch
- Willingness to train large deep learning models on GPU-based systems
Nice to Have
- ML model design, optimization, and HW-SW co-development methodology
- ML model optimization targeting PPA on Arm compute platforms
- Adaptability to fast-paced ML industry and willingness to learn new technologies
Benefits
- Competitive salary ($50-60/hour)
- On-the-job learning, mentoring, and buddy schemes
- Regular feedback, development opportunities, social activities
- End of internship celebration
- Consideration for future graduate positions (subject to performance)