Role overview

Scale is seeking Machine Learning Research Interns for Summer 2027 to join a team building the foundation for next-generation AI systems. The company develops high-quality data and full-stack technologies that power leading models, working with industry leaders and government agencies to build, deploy, and oversee AI applications. As an intern, you’ll contribute to real, published or shipped work alongside world-class researchers rather than side projects.

Key responsibilities

Measure dangerous capabilities of frontier models and conduct preparedness research.
Research and develop new benchmarks for evaluating frontier models.
Create novel training methods to help models solve extremely difficult reasoning problems requiring extended chains of thought.
Investigate reinforcement learning and post-training approaches that enhance model reliability on high-stakes tasks.
Explore scalable oversight protocols enabling humans to produce and quality control reasoning chains beyond their native capabilities.
Analyze model generalization boundaries and capabilities to inform data-driven improvements.
Work on synthetic and hybrid data generation with human input to scale high-quality data production.
Improve production models through retraining and hyperparameter optimization, then deploy while maintaining core model performance.
Develop post-training or agentic solutions that integrate into enterprise and public sector applications.

Requirements

Currently enrolled in a BS, MS, or PhD program in Machine Learning, Deep Learning, or Natural Language Processing with a graduation date in Fall 2027 or Spring 2028.
Available to start in May or June 2027 in either San Francisco or New York.
Prior research experience or publications on large language models, NLP, multimodal models, agents, safety, evaluation, alignment, or related fields.
Proficiency with one or more general-purpose programming languages such as Python or Javascript.

What makes this role worth considering

This internship offers exposure to cutting-edge research spanning generative AI, reinforcement learning, scalable oversight, and evaluation science. You’ll work on a large-scale hybrid human-machine system supporting dozens of industry-leading customers and processing billions of tasks monthly, giving you insight into some of the most complex AI use cases in the industry.