Sr Applied Scientist, ML Codesign, Edge AI Platform
Core
Define joint optimization of model compression and silicon architecture for Amazon's next-generation edge and cloud inference accelerators.
Role type
Sr. Applied Scientist (ML Codesign, Edge AI Platform)
Builds
Next-generation ASICs and compression algorithms for edge/cloud inference
Domain
Edge AI, Machine Learning, Silicon Architecture
Deliverable
production ML models
Required skills
Machine learning model building, neural deep learning methods, Java, C++, Python, post-training quantization, quantization-aware training, knowledge distillation, structured pruning, hardware-aware training, benchmark validation (MMLU, GSM8K, HumanEval, IFEval), team mentorship
Preferred skills
R, scikit-learn, Spark MLLib, MxNet, Tensorflow, numpy, scipy, large scale distributed systems (Hadoop, Spark)
Technologies
Java, C++, Python, Tensorflow, numpy, scipy, Hadoop, Spark, MxNet, scikit-learn, R
Responsibilities
Define hardware-aware compression roadmap, own joint optimization of compression algorithms with hardware, represent applied science in silicon architecture reviews, set science roadmap for compression techniques, mentor applied scientists, serve as single-threaded technical leader for codesign agenda
Seniority
Senior, hands-on IC with mentorship
