At Crowdgen / AI Data Projects, I evaluated LLM-generated responses for accuracy, relevance, instruction following, and contextual quality. I also annotated text data and checked datasets for errors, inconsistencies, and labeling issues.
With SAMA-AI / Speech Data Projects, I analyzed speech and audio datasets, transcribed speech, and reviewed transcripts for accuracy and linguistic consistency. I applied knowledge of African-language speech and cultural context when evaluating the data.
I bring native/local understanding of Kikuyu / Gikuyu and familiarity with Swahili, Kenyan English, regional language variation, and code-switching. My background as a Registered Nurse / Mental Health Nurse also involved accurate documentation and handling confidential information.

