Job Description

Labelbox is seeking an Applied Research Engineer to develop cutting-edge systems and methods for creating, analyzing, and leveraging high-quality human-in-the-loop data for frontier model developers. The role involves designing and implementing systems for AI training processes like RLHF and DPO, improving data quality, and developing AI-assisted tools for data labeling.

Responsibilities

• Develop AI alignment methods like RLHF
• Improve human-in-the-loop data quality
• Create AI-assisted data labeling tools
• Investigate impact of human feedback on model performance
• Optimize human feedback collection algorithms
• Integrate breakthroughs into Labelbox’s product suite
• Engage with customers and the AI community
• Publish research in top-tier journals
• Stay ahead of AI advancements
• Create technical documentation and educational content

Requirements

• Ph.D. or Master’s degree in Computer Science, Machine Learning, or related field
• 3+ years of experience solving complex ML challenges
• Expertise in data quality measurement and refinement
• Deep understanding of frontier AI models
• Proficiency in Python and deep learning frameworks (PyTorch, JAX, TensorFlow)
• Track record of publishing in top-tier AI/ML conferences
• Ability to bridge research and application
• Strong analytical and problem-solving skills
• Exceptional communication and collaboration skills

Benefits

• No benefits