AI Trainer (Freelance), Outlier AI
OtherText
As an AI Trainer (Freelance) at Outlier AI, I designed research prompts to challenge large language models on finance topics. My tasks included creating adversarial prompts, evaluating responses for accuracy, and verifying AI-generated content against regulatory standards. This required applying financial expertise to ensure model outputs met strict analytical and factual thresholds.
• Designed multi-step evaluation rubrics for LLMs specializing in financial reasoning
• Conducted adversarial testing and identified model reasoning failures
• Verified outputs using regulatory and industry-specific data
• Evaluated and scored LLM responses against multiple metric dimensions.
As an AI Trainer (Freelance) at Outlier AI, I designed research prompts to challenge large language models on finance topics. My tasks included creating adversarial prompts, evaluating responses for accuracy, and verifying AI-generated content against regulatory standards. This required applying financial expertise to ensure model outputs met strict analytical and factual thresholds. • Designed multi-step evaluation rubrics for LLMs specializing in financial reasoning • Conducted adversarial testing and identified model reasoning failures • Verified outputs using regulatory and industry-specific data • Evaluated and scored LLM responses against multiple metric dimensions.
2026 - Present