AI & Frontend Engineer (AI Trainer/Evaluator)
As an AI & Frontend Engineer at Yfeey — AI Agency, I evaluated large language model outputs for accuracy, tone, and safety to enhance system reliability. I crafted and iterated on prompts to optimize LLM responses and documented instances of failure patterns for future improvements. My work ensured that responses from integrated AI models were appropriate, relevant, and aligned with client requirements. • Assessed and rated text-based AI model outputs against quality metrics • Refined prompts based on observed inaccuracies and safety risks • Utilized OpenAI API, Claude API, and Mastra AI SDK in production environments • Documented recurrent issues to inform prompt and workflow improvements.