Join the Frontier Data Research team to discover new forms of supervision at the bleeding edge of advanced reasoning models. On the team, you will study novel data formats, supervision methods, and post-training algorithms to improve the reasoning capabilities of large language models. Candidates with backgrounds and interest in developing models to do expert level math, software engineering, biology, physics, and chemistry are particularly encouraged to apply.
Example Projects:
- Research and develop new methods for training models to excel on extremely difficult reasoning problems that require long chains of thought
- Research scalable oversight protocols that enable humans to produce and quality control reasoning chains beyond their native capabilities
- Investigate strategies to refine and enhance data pipelines for model improvement
- Study the boundaries of reasoning model generalization to inform data-driven advancements
- Develop new benchmarks or evaluation methods for gauging advanced reasoning capabilities via outcome or process supervision
Required to have:
- Currently enrolled in a BS/MS/PhD Program with a focus on Machine Learning, Deep Learning, Natural Language Processing, or Computer Vision with a graduation date in Fall 2025 or Spring 2026
- Prior work experience or track record of research publications on LLMs, reasoning, NLP, agents, alignment or a related field
- Experience training neural networks in Python
- Ability to speak and write in English fluently
- Be available for a Summer 2025 (May/June starts) internship
Ideally you’d have:
- Have had a previous internship involving LLMs, Machine Learning, Deep Learning, Natural Language Processing, Alignment, or Agents.
- Experience as a researcher, including internships, full-time, or at a lab
- Publications in top-tier ML conferences such as NeurIPS, ICLR, CVPR, ICML, COLM, etc. or contributions to open-source projects
- Screaming good coding capabilities as evidenced by open source projects
