Research Engineer - Data Infrastructure

What you'll do

• Build pipelines to collect, process, filter, and transform training datasets.
• Train models for data processing, including classifiers, quality filters, and labeling models; develop curation strategies to improve model performance.
• Create tools and infrastructure for researchers to explore and train on massive datasets quickly and reliably.

What they ask for

• Experience building data-intensive systems, ideally for machine-learning training pipelines.
• Strong engineering skills in distributed data processing at scale, such as Kubernetes or custom pipelines over large datasets.
• Ability to evaluate how data quality, composition, and curation affect model outcomes and build tools to measure this.

Nice to have

• Experience building or operating web crawlers.

Tools

Kubernetes

AI focus

Develop data infrastructure and curation methods for frontier AI models, including training classifiers, quality filters, and labeling models for data-processing pipelines.