Meritt North

I am a creative, detail-oriented AI trainer and content specialist with extensive experience in data annotation, model evaluation, and multimodal content analysis. I train and evaluate AI and LLM systems through detailed text, audio, and visual data annotation, and I assess outputs for accuracy, clarity, bias, and contextual relevance.

I produce and review high-quality transcription and speech data for voice recognition and NLP models, supporting audio segmentation and emotion/tone labeling for human-in-the-loop multimodal training sets. I also manage end-to-end client projects emphasizing storytelling, clarity, and consistent brand voice across podcasts, audiobooks, and e-learning materials.

Skilled in transcription, prompt creation, structured dataset development, and QA/dataset curation, I use tools such as Descript, Reaper, Audacity, Auphonic, Adobe Audition, and AI systems like ChatGPT, Whisper, Claude, and Hugging Face to improve model reasoning, fluency, and contextual understanding.