High Noon
In High Noon we developed rubrics to assist the evaluation of two o three model responses.
Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
No subject matter listed
I’ve worked on AI training-data projects where the goal isn’t just “labeling”, but producing high-fidelity, auditable judgments that improve model behavior. My work has included rubric-based evaluation of model outputs (accuracy, faithfulness to instructions, calibration or overclaiming, tone and safety constraints), as well as writing and refining prompts and test cases that stress models under ambiguity, cross-dependencies, and edge conditions. What sets me apart is that I bring a quantitative methods mindset to training data: I treat annotation as measurement. I can run lightweight QA in R/Python to spot drift, inconsistencies, and systematic error patterns. I’m also strong at structured writing, so I can produce clear, reproducible rationales without verbosity.
In High Noon we developed rubrics to assist the evaluation of two o three model responses.
Prompting and model evaluation based on PhD-level rubrics.
PhD, Public Policy
MA, Public Policy
Researcher
Deputy Director General for Research