AI Workflow Developer—LLM Prompt Engineering and Evaluation
I designed and refined prompt engineering pipelines, working directly with LLM outputs to ensure accuracy and consistency. My workflow included hands-on LLM evaluation, structured SOP development, and prompt optimization for automated AI systems. I collaborated on practical, user-facing data training and evaluation workflows relevant to RLHF and model training. • Developed structured prompt-response tasks for fitness plan generation using n8n and DeepSeek API • Evaluated and rated LLM outputs based on fitness data and user requirements • Built and maintained automated procedures for AI workflow evaluation and prompt iteration • Ensured longitudinal data consistency by implementing robust data pipelines