AI Trainer / Data Contributor (Remote) — Mindrift AI
Trained and evaluated AI models on mathematics problem content spanning algebra, calculus, probability, geometry, and statistics. Assessed model responses for logic, numerical accuracy, and sequential correctness, then used findings to improve model behavior. Developed high-quality quantitative prompts and tested datasets containing mathematical equations, formulas, and statistical results. • Produced/train prompts using quantitative reasoning • Evaluated answers for correctness and reasoning order • Tested STEM datasets with equations and statistical outputs • Identified reasoning errors to support model improvements