AIML Researcher/Engineer - Foundation Model Post-Training

What you'll do

• Design end-to-end post-training strategies, including reinforcement learning, and develop preference optimization, steering, and safety algorithms.
• Research human and synthetic data generation, filtering, curriculum learning, and evaluation methods for model performance.
• Collaborate with pre-training and product teams on architecture choices and translating user needs into model capabilities.

What they ask for

• Expertise in deep learning focused on LLMs, post-training, or reinforcement learning, supported by academic or practical accomplishments.
• Programming proficiency in Python and a major deep learning framework such as JAX or PyTorch.
• Master's/PhD, or equivalent practical experience, in Computer Science, Machine Learning, or a related technical field.

Nice to have

• Experience training state-of-the-art large models at scale, including distributed training challenges.
• Experience improving complex reasoning performance or working with transformer architectures and their transformations.

Pay

USD $184,700 to $324,800 base salary; USD $175,000 to $308,500 base salary

Tools

Python, JAX, PyTorch

AI focus

Foundation-model post-training, including reinforcement learning, instruction following, tool use, reasoning, preference optimization, model steering, safety, data strategy, and evaluation.