Digital Content Manager (AI/LLM Content Evaluation) - Cobalt Intelligence
Led B2B fintech and alternative-lending content production for AI evaluation workflows using prompt engineering and rubric-based quality scoring. Developed evaluation systems to compare AI-generated drafts against quality criteria and reviewer personas before publication. Supported factual accuracy, tone, and domain correctness checks while building Python tooling to automate verification and content workflows. • Engineered prompts for LLMs such as Claude and GPT • Built FastAPI Python tools for data verification and automation • Created rubric-based evaluation and RLHF-style preference review processes • Reviewed model outputs for compliance, accuracy, and domain correctness