Obsidian seeks a domain-expert engineer for a frontier engineering-reasoning evaluation with a leading AI research lab. You explore whether state-of-the-art models can reason from first principles in engineering domains and reveal the model's internal reasoning on tasks.
In practice you define the simulation, set specs with pass/fail thresholds, and craft the prompts. The model proposes a design and an agentic grader scores satisfaction against your criteria.
#J-18808-LjbffrControl Systems Architect for AI-Evaluated Design in new york at Unknown Company
This position is listed as full time and onsite.