Research Engineer, Data Foundations

This is a fully remote position, open to applicants in United States.

📋 Description

• Develop multimodal, multitask datasets that enable world models to acquire new capabilities — determining the data to collect, generate, or curate and assessing its impact on model behavior.

• Conduct controlled training experiments to analyze how data composition influences model performance across various tasks and domains.

• Create and maintain large-scale pipelines for synthetic data generation, filtering, and quality assurance.

• Establish evaluations and benchmarks to determine if our models are genuinely improving in critical areas.

• Collaborate with product and creative teams to convert target behaviors and capabilities into actionable data strategies.

⛳️ Requirements

• A minimum of 4 years of experience in machine learning, with additional preference for data-centric methodologies.

• Familiarity with large multimodal datasets and generative models (video, image, or multimodal).

• Strong intuition regarding how data composition and quality relate to model capabilities.

• Ability to work across the entire research stack: data analysis, dataset creation, model training, evaluation, and iterative improvement.

• Expertise in at least one machine learning framework (e.g., PyTorch, JAX) and distributed computing tools (e.g., Ray, Kubernetes).

• Passion for developing AI that effectively simulates the world.

🏝️ Benefits

• Health insurance

• 401(k)

• Flexible work arrangements

• Paid time off