Lead AI Automation Consultant – RLHF & Data Annotation
I architected RLHF pipelines and conducted high-fidelity data annotation for large language models (LLMs) with a focus on grounding them in East African contexts. My responsibilities included prompt engineering, red-teaming for safety and bias, and instruction tuning through detailed human-in-the-loop feedback frameworks. I ensured data integrity and logical consistency in model training, emphasizing cross-cultural nuances for regional precision. • Developed workflow automations for RLHF and data annotation tasks. • Conducted safety and bias evaluation on multilingual LLMs via red-teaming exercises. • Delivered comprehensive feedback for LLM instruction tuning in African domains. • Leveraged tools such as Label Studio and Google Cloud Vertex AI for high-volume annotation.