Advanced Subject Matter Expert (SME) & AI Risk Partner at Scaled AI / Outlier Systems
Provided AI risk and safety alignment evaluation support for advanced language models in a financial mathematics and compliance-focused context. Authored targeted reference datasets with explicit step-by-step reasoning constraints and alignment guidelines to improve model deployment safety. Audited multi-turn logic trees and verification scripts to identify mathematical regressions, computational holes, and edge-case errors. • Conducted systematic logic evaluations and edge-case stress testing for compliance quality • Produced and maintained evaluation reference datasets for ranking within premium tier brackets • Performed audits of reasoning constraints and verification scripts for consistency • Supported complex institutional risk simulation streams with safety alignment oversight