Generalist AI Trainer
As a Generalist AI Trainer, I evaluated and ranked AI-generated responses in various domains including finance, business, economics, and general knowledge. I wrote and refined prompts to test and improve model reasoning, and provided detailed annotations and rationales to inform reinforcement learning from human feedback (RLHF) pipelines. My work specifically leveraged domain expertise in finance and M&A to ensure technical accuracy in specialist outputs. • Evaluated and ranked outputs of large language models for accuracy, coherence, and helpfulness. • Crafted and edited prompts to probe model capabilities and surface edge-case behaviors. • Provided detailed preference annotations and written rationales for RLHF processes. • Applied subject matter expertise in finance to assess and guide specialist model outputs.