Moses Ogero

I am an analytical, detail-oriented RLHF and data annotation specialist focused on improving LLM accuracy, safety, and helpfulness. I concentrate on Human-in-the-Loop workflows and model evaluation to reduce bias and hallucinations.

I have a consistent track record of maintaining 98%+ accuracy in high-volume, complex annotation environments and enforce 100% adherence to evolving project taxonomies. I perform deep-dive evaluations of AI-generated content with emphasis on factual verification and logical consistency.

I collaborate with project leads to refine labeling guidelines and developed a systematic peer-review approach that reduced departmental error rates by 15%. I am skilled at synthesizing dense project instructions to minimize model error and spot rare data anomalies.

My technical toolkit includes semantic analysis, sentiment labeling, prompt engineering, Markdown and JSON formatting, Labelbox/Scale/Appen annotation platforms, Microsoft Excel (advanced), Google Workspace, Slack, Trello, and basic Python for data cleaning. I reliably work 100% remote.