Amazon Devices is looking to shape the next generation of ML-enabled consumer devices by developing a new architecture powering Amazon's Neural Edge Engine and building custom accelerators optimized for deep learning workloads.
Requirements
- 3+ years building ML models for real-world applications
- Strong programming experience in Java, C++, Python, or similar
- Deep understanding of neural networks and machine learning
- Experience with modelling tools such as R, scikit-learn, Spark MLlib, MXNet, TensorFlow, NumPy, SciPy
- Experience with large-scale distributed systems (Hadoop, Spark, etc.)
Responsibilities
- Analyse deep learning workloads and map them to Amazon’s Neural Edge Engine
- Propose and implement new hardware architectures and improvements for future ML workloads
- Collaborate with compiler engineers, model developers, architects, and product teams to deliver ML-centric hardware/software solutions
- Deliver hardware architecture, microarchitecture, and design collateral for next-gen ML accelerators
- Build tools for performance modelling and evaluation across power, performance, and cost trade-offs
- Work with silicon design teams to bring architectures from concept to silicon
Other
- PhD, or Master’s degree with 6+ years of applied ML research
