Senior Applied Scientist, AGI Customization
Core
Developing state-of-the-art services and tools for customizing large language models (Amazon Nova) to enable enterprises to build application-specific models without training from scratch.
Role type
Senior Applied Scientist (LLM Customization)
Builds
Enterprise-ready tooling for model customization techniques (SFT, DPO, PPO) on Amazon SageMaker
Domain
Artificial General Intelligence (AGI), Large Language Models, Machine Learning
Deliverable
production ML models
Required skills
supervised fine-tuning, reinforcement learning, knowledge distillation, neural deep learning methods, experiment design, preference learning algorithms, evaluation frameworks, secure access mechanisms
Preferred skills
distributed systems (Hadoop, Spark), modeling tools (Tensorflow, numpy, scipy)
Responsibilities
Design and execute experiments to optimize model accuracy, latency, and cost; Develop and enhance preference learning algorithms and training curricula; Create robust evaluation frameworks for assessing model performance; Contribute to the development of the Responsible AI toolkit; Design and implement secure access mechanisms for early model checkpoints and weights
Seniority
Senior, hands-on IC