Emmanuel Oginni

My work focuses on improving the quality and reliability of AI systems through structured human evaluation.
Over the past few years, I've worked across multimodal AI projects involving image, video, text, and audio data. My experience includes evaluating AI-generated outputs, reviewing image transformations, performing OCR verification, analyzing video content, validating annotations against project guidelines, and identifying inconsistencies that affect model performance.
Alongside my industry experience, I recently completed a Master's in Information Management at Ahmadu Bello University. My research explored how information management practices influence data quality, governance, and security principles that closely align with the work required to build reliable AI systems.
I enjoy work that requires careful observation, structured reasoning, and evidence-based decision-making. Whether reviewing model outputs, validating annotations, or documenting edge cases, my goal is always the same: produce consistent, trustworthy evaluations that improve data quality and support better AI models.
Areas I work in:
• Multimodal AI Evaluation (Image, Video, Text & Audio)
• AI Quality Assurance
• Human Feedback & Model Evaluation
• Image Transformation Review
• OCR Verification
• Frame-level Video Analysis
• Annotation Quality Review
• Guideline Compliance
• Hallucination & Factuality Assessment
• Data Validation & Information Quality
I'm always interested in opportunities involving AI evaluation, multimodal quality assurance, RLHF-style evaluation, and annotation workflow improvement.