Member of Technical Staff, Research Engineer (Datasets)

About the role

Building general world models — systems that understand and simulate reality across tasks, modalities, and domains — demands training data that is as rich and varied as the real world itself. We’re looking for Research Engineers to own the data behind our models: what they learn from, how well they learn it, and what new capabilities that unlocks. You will design datasets, run modeling experiments, and build the infrastructure to generate and curate data at scale — directly shaping what our models can do, with applications ranging from creative tools to robotics.

What you'll do

• Design multimodal, multitask datasets that teach world models new capabilities — deciding what data to collect, generate, or curate and measuring its effect on model behavior
• Run controlled training experiments to understand how data composition drives model performance across tasks and domains
• Build and operate large-scale pipelines for synthetic data generation, filtering, and quality control
• Define evaluations and benchmarks that measure whether our models are actually improving at the things that matter
• Partner with product and creative teams to translate target behaviors and capabilities into concrete data strategies

What you'll need

• 4+ years of experience in machine learning, bonus points for data-centric approaches
• Experience with large multimodal datasets and generative models (video, image, or multimodal)
• Deep intuition for how data composition and quality translate to model capabilities
• Comfort working across the full research stack: data analysis, dataset creation, model training, evaluation, and back again
• Proficiency with at least one ML framework (e.g. PyTorch, JAX) and distributed compute tools (e.g. Ray, Kubernetes)
• Excitement about building AI that simulates the world