Member of Technical Staff - Pre-Training Data
Core
Design and implement techniques for sourcing, curating, and filtering large-scale web, text, and code data for open AI models.
Role type
Pre-training data researcher
Builds
Open foundational models
Domain
Artificial Intelligence / Machine Learning
Deliverable
production ML models
Required skills
Python, deep learning frameworks (PyTorch/JAX), distributed computing, data curation, data quality metrics, experimental design, ablation studies, data ethics
Preferred skills
Probability and statistics, large-scale text/code dataset analysis, data engineering, data safety and licensing frameworks, open source contributions, PhD in CS/ML/Physics/Mathematics
Technologies
Python, PyTorch, JAX
Responsibilities
Design and run data experiments, develop data quality metrics, mitigate data risks (privacy/safety/licensing), evaluate dataset changes via ablations, publish research and share code/datasets
Seniority
Mid-level IC (Bachelor's required, PhD preferred)
