Research Scientist, AI Benchmarks

Job Summary

To support the development of AI benchmarks, the full-time remote Research Scientist, AI Benchmarks will design public benchmarks for image, video, and document models, build private test sets from real data, and run data-value experiments to measure the impact of Kled data on model performance.

Key responsibilities

• Design and ship public benchmarks for various AI models
• Build private test sets using real, opt-in data
• Run experiments to evaluate the value of Kled data in training models

Required qualifications

• 5-7+ years of ML research or research engineering experience
• Proven experience creating benchmarks or evaluations used by others
• Strong hands-on experience with training and fine-tuning image or video models using PyTorch
• Deep understanding of experimental design and statistics
• Experience running end-to-end experiments from GPUs to final reporting