Research Scientist, Mid-training - Synthetic Data
Description For Research Scientist, Mid-training - Synthetic Data
The Mid-training team at OpenAI is seeking a Research Scientist for Synthetic Data to improve their flagship models. This role involves creating synthetic data, conducting research on synthetic data methods, and using it to train state-of-the-art models. The ideal candidate should have a strong technical background, familiarity with the LLM stack, and experience in applied Machine Learning. Key responsibilities include generating different types of synthetic data, training models, running experiments, and leading research initiatives involving data quality. The position requires 3+ years of experience in machine learning, proficiency in Python, and familiarity with Deep Learning and Large Language Models. Experience with synthetic data or leading large LLM training efforts is a bonus. This San Francisco-based role offers a hybrid work model and the opportunity to contribute to cutting-edge AI research and development at OpenAI.
Responsibilities For Research Scientist, Mid-training - Synthetic Data
- Generate different types of synthetic data
- Train models and run experiments to study the quality of synthetic data
- Train SOTA models using the generated synthetic data
- Lead other research bets involving data and data quality
Requirements For Research Scientist, Mid-training - Synthetic Data
- 3+ years of experience working with machine learning
- Highly skilled with Python
- Capable of leading ambitious machine learning projects
- Familiarity with Machine Learning, Deep Learning, and Large Language Models and related infrastructure (e.g. PyTorch)
Benefits For Research Scientist, Mid-training - Synthetic Data
- Relocation assistance
