• Author complex, high-fidelity reasoning traces that capture planning, tool use, and multi-step decision-making for sophisticated technical tasks
• Design and document structured traces that demonstrate how a model should reason through real-world scenarios — not just what the answer is, but how to get there
• Review and quality-check traces produced by other contributors, ensuring clarity, logical soundness, and technical accuracy
• Develop data strategies that help LLMs navigate ambiguous, multi-layered problems with reliable, consistent reasoning
• Apply senior-level architectural insight to ensure traces reflect best practices in model decision-making and problem decomposition

Who You Are

• Experienced in machine learning, AI research, or a closely related technical field — you understand how models learn and where they fail
• Skilled at decomposing complex problems into clear, documented, logical steps that others can follow and evaluate
• Familiar with LLM training pipelines, evaluation methodologies, and the mechanics of how models develop reasoning capabilities
• A precise, structured thinker who can translate expertise into high-quality written documentation
• Self-directed and reliable — you produce consistent, high-quality work without hand-holding

Nice to Have

• Prior experience with data annotation, data quality assurance, or AI evaluation systems
• Top-tier Kaggle competition results (Grandmaster or Master level) demonstrating elite-level model understanding and performance optimization
• Background in reinforcement learning, chain-of-thought prompting, or agentic AI systems
• Experience working with or evaluating frontier LLMs

Why Join Us

• Work directly with some of the world's leading AI research teams and labs on cutting-edge model development
• Fully remote and flexible — structure your work around your life, not the other way around
• Freelance autonomy with meaningful, intellectually stimulating work on problems that actually matter
• Gain deep, hands-on exposure to how frontier LLMs are trained and evaluated
• Potential for ongoing work and contract extension as new projects launch