Peter Abok

I am a versatile, detail-oriented AI Training Specialist with over three years delivering high-accuracy annotation across CV, NLP, and generative AI domains.

I specialize in video, audio and text processing including segmentation, motion tracking, linguistic labeling and prompt engineering, and have achieved 99.5% accuracy across 50,000+ data points.

My work has boosted model performance 15–30%, cut project timelines by 30% through tool optimization, and improved captioning/model accuracy via meticulous labeling and QA protocols.

I collaborate closely with engineering and design teams, mentor junior annotators, and implement standardized evaluation protocols and A/B testing methods to produce scalable, high-fidelity training data.