Ayodele Ogunjemite

At Appen, I lead 12 annotators across image classification, NLP, and audio transcription projects, maintaining quality scores above 97% across active workstreams.

I define annotation guidelines, calibration protocols, and edge-case escalation paths for LLM training datasets. I also use Cohen's Kappa for inter-annotator agreement analysis and partner with ML engineers to align labelling schemas with evolving model requirements.

Through Outlier AI/Aether, I evaluated LLM responses for factual accuracy, tone, instruction-following, and safety, specialising in RLHF preference ranking. I maintained 90+ tasks daily while ranking in the top 10% of the annotator pool for quality.

My experience also includes research, workforce, and healthcare analytics at Andela, General Hospital Ikorodu, and Data Science Nigeria, where I cleaned data, built dashboards, structured qualitative datasets, and contributed to Yoruba and Pidgin English NLP datasets.