AIML Researcher/Engineer - Foundation Model Post-Training

What you'll do

• Design and refine post-training strategies to enhance model behaviors using reinforcement learning techniques.
• Develop innovative algorithms for optimizing preferences, steering models, and ensuring safety in AI systems.
• Lead data strategy initiatives by researching methods for generating high-quality human and synthetic training datasets.
• Create robust evaluation frameworks to measure model performance in helpfulness, factuality, and utility beyond static benchmarks.
• Collaborate with pre-training teams to guide architectural decisions that align with user-centric requirements.

What we're looking for

• Demonstrated expertise in deep learning, LLMs, post-training, or reinforcement learning.
• Proficient programming skills in Python and major deep learning frameworks like JAX or PyTorch.
• Masters/PhD or equivalent experience in Computer Science, Machine Learning, or related field.
• Experience training large-scale models with distributed training challenges knowledge.
• Strong communication skills for cross-functional collaboration across research and product teams.