Beth Mwangi

Detail-oriented AI annotation and research assistant with hands-on experience collaborating with AI systems to produce accurate, well-reasoned, and clearly written outputs. Experienced in video annotation, data labeling, content evaluation, and guideline-driven analysis through Atlas Capture. Strong web research, fact-checking, and English writing skills, with a growing focus on human-AI collaboration, prompt refinement, and AI output quality assurance.

Comfortable adapting quickly across diverse tasks and working alongside large language models to deliver reliable, real-world-ready results.

AI Content Annotator / Video Annotation Specialist at Atlas Capture (2024-01 – Present)

Performed video annotation and data labeling tasks to support AI and machine learning model training. Collaborated indirectly with AI systems by reviewing, correcting, and refining model-assisted outputs. Applied goal-oriented segmentation principles to label complex task sequences accurately.

• Performed video annotation and data labeling tasks to support AI and machine learning model training
• Collaborated indirectly with AI systems by reviewing, correcting, and refining model-assisted outputs
• Applied goal-oriented segmentation principles to label complex task sequences accurately
• Evaluated multimedia content including social media videos, instructional clips, and documentary-style footage
• Followed detailed project guidelines to ensure accuracy, consistency, and reliability of training data
• Flagged ambiguous, incorrect, or low-quality outputs to improve downstream AI performance
• Demonstrated strong attention to detail while working independently in a flexible, task-based environment
• Delivered high-quality annotations with minimal revisions
• Trusted with guideline-heavy tasks requiring judgment and precision
• Maintained quality standards across changing project requirements

Bachelor's Degree in Multimedia – Multimedia University of Kenya (2022)