Ben Feingold

I spent over a decade in trade credit and financial analysis, where I learned to spot inconsistencies in complex data, assess risk under pressure, and make clear judgements in ambiguous situations. More recently, I've shifted to AI model evaluation and annotation, working with Mercor Intelligence on different projects where I assess AI outputs and grade them for accuracy across text, visuals, and audio.

My core skill is evaluating whether systems — financial or AI-powered — are making sound decisions. I review structured and unstructured data, verify accuracy, identify edge cases, and feed findings back to improve system performance. I'm experienced in data annotation, AI output validation, quality assurance, and the kind of methodical human oversight that keeps training pipelines honest.