AI Trainer (Prompt Engineer & Output Evaluator)
I engineered multi-turn prompt chains and evaluated AI outputs to enhance the capabilities of autonomous LLM agents. My work involved crafting, reviewing, and documenting high-quality training data for AI models, with a focus on effective prompt engineering and real-world automation tasks. I contributed technical expertise in coding and multilingual support, ensuring the relevancy and accuracy of training materials. • Built and refined prompt+response datasets for LLM agents using Groq API (LLaMA3-70b). • Designed evaluation criteria and rated responses for correctness and context appropriateness. • Applied hands-on knowledge in coding, trading analysis, and language localization (EN/FR/RW). • Ensured data quality through version control and meticulous documentation.