Elijah Johnson

I’m a detail-oriented AI Data Training & Evaluation Specialist with a strong analytical foundation from a B.S. in Mathematics. I focus on improving AI model accuracy, consistency, and reliability through structured dataset review, evaluation, and quality scoring—especially for multimodal work across text, audio, and visual content.

In my Handshake AI projects, I evaluated AI-generated images for prompt alignment, accuracy, and visual quality, using criteria like correctness, naturalness, and instruction following. I also trained and refined multimodal systems by evaluating and annotating datasets, maintaining required baseline quality scores, and flagging data inconsistencies with structured feedback to reduce error rates and strengthen large-scale training pipelines.