Research, Pre-Training Data

What you'll do

• Design techniques to source, curate, and filter large-scale text, code, and multimodal data; measure dataset quality, coverage, diversity, and representativeness.
• Scale reproducible data processing with research and infrastructure teams, and assess how dataset changes affect model learning and behavior.
• Investigate privacy, safety, and licensing risks; publish and present research and share code, datasets, and insights.

What they ask for

• Proficiency in Python and familiarity with at least one deep learning framework.
• Ability to debug distributed training and write scalable code.
• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline, with strong theoretical and empirical grounding.
• Clear written communication and ability to explain complex technical concepts.

Nice to have

• Experience with large-scale dataset curation, preprocessing, analysis, data engineering, dataset construction, or web data processing for machine learning.
• Experience evaluating training data quality, knowledge of data ethics, safety, and licensing frameworks, or contributions to open datasets, publications, or data tooling.

Pay

$350,000 - $475,000 USD (annual base salary)

Tools

Python, PyTorch, TensorFlow, JAX

AI focus

Develop and evaluate pre-training datasets and data systems for frontier AI models, including responsible data curation and analysis of downstream effects on model learning and behavior.