NVIDIA seeks a senior software engineer to define and evolve the technical approach for evaluating and improving open agent harnesses across frameworks, models, and benchmarks. You will build scalable evaluation systems, diagnose failures from traces and logs, and design experiments to distinguish real harness improvements from noise.
Responsibilities include shipping harness/runtime improvements, contributing to open-source, and collaborating with model, infrastructure, and research teams to
#J-18808-LjbffrSenior AI Agent Evaluation Engineer in california at Unknown Company
This position is listed as full time and onsite.