Valkyrie – AI Model Evaluation and Prompt Development
OtherDocumentQuestion AnsweringText Generation
Ongoing AI training project involving evaluation and improvement of LLM responses to advanced mathematical and logic-based prompts. Tasks include creating prompts and rubrics, rating AI outputs for truthfulness, completeness, and clarity, and writing model-quality completions for supervised fine-tuning (SFT). Ensured consistency with detailed guidelines, applying structured reasoning and accuracy-focused review to thousands of samples. All work undergoes quality assurance checks to maintain alignment standards and model improvement goals.
Ongoing AI training project involving evaluation and improvement of LLM responses to advanced mathematical and logic-based prompts. Tasks include creating prompts and rubrics, rating AI outputs for truthfulness, completeness, and clarity, and writing model-quality completions for supervised fine-tuning (SFT). Ensured consistency with detailed guidelines, applying structured reasoning and accuracy-focused review to thousands of samples. All work undergoes quality assurance checks to maintain alignment standards and model improvement goals.
2025