I have contributed to AI training and evaluation, as well as data labelling across three platforms, which include Mercor
I have contributed to AI training and evaluation, as well as data labelling across three platforms, which include Mercor, Micro1, and Verita AI, and holding roles as Finance Expert, Corporate Finance Expert, FP&A Expert, Professional Domain Expert (Finance & Accounting), and Generalist. My work spans the full training pipeline, including prompting, task creation, task review, ground truth file development, model preference labelling, model review, grader feedback, and app review and comparison. As a domain expert, I apply over 13 years of senior finance and strategy experience to evaluate and shape how AI models reason through complex financial problems. This includes assessing outputs across FP&A, corporate finance, accounting, and broader business finance contexts, judging technical accuracy, logical coherence, and professional utility against the standard a qualified finance professional would expect. In my Generalist capacity, I extend this rigour to cross-domain tasks, bringing the same structured analytical lens to a wider range of prompts and responses. Across these engagements, I have developed a strong understanding of how large language models handle financial reasoning, where they fall short, and how task design and evaluation quality directly influence model behaviour.