Key Responsibilities
- Author high-fidelity reasoning traces that capture planning, tool use, and multi-step decision-making for complex technical tasks
- Design and document structured traces demonstrating how models should reason through real-world scenarios
- Review and quality-check traces produced by contributors, ensuring clarity, logical soundness, and technical accuracy
- Develop data strategies to help LLMs navigate ambiguous, multi-layered problems with reliable reasoning
- Apply senior-level architectural insight to ensure traces reflect best practices in model decision-making
- Decompose complex problems into clear, documented logical steps for model evaluation
Requirements
- Deep expertise in machine learning, AI research, or related technical fields with understanding of model behavior
- Proven ability to decompose complex problems into structured, logical steps
- Familiarity with LLM training pipelines, evaluation methodologies, and model reasoning mechanics
- Strong written communication skills for creating high-quality technical documentation
- Self-directed work ethic with consistent output quality