At OneForma, I evaluate and rank AI-generated responses for relevance, accuracy, helpfulness, coherence, and instruction adherence. I also provide structured human feedback for LLM training and grade AI safety and content according to project guidelines.
At LinguaTech Analytics, I annotated and transcribed over 600 hours of English, French, and Igbo voice data, contributing to a 22% improvement in voice recognition accuracy. I also developed translation consistency templates that reduced post-editing time by 30%.
