Stanslaus Bosire

I’m a detail-oriented AI Trainer and Data Annotator with 2+ years of hands-on experience supporting AI system development through precise data labeling, categorization, and quality review. I consistently synthesize large volumes of text, image, and LiDAR data into structured, model-ready outputs that improve accuracy and reliability.

I apply predefined rubrics, taxonomies, and annotation guidelines with precision, including Blend’s exact-match decision framework. I’m known for flagging dataset inconsistencies, ambiguities, and errors—escalating appropriately and contributing to continuous annotation process improvement through clearer documentation and updated guidance.

In my recent role as a Night Shift Supervisor at Digital Divide Data Kenya (DDD), I supervise annotation team operations to meet daily quality benchmarks and project SLAs. I monitor annotator performance against rubric standards, coordinate task allocation, and help refine guidelines based on recurring error trends—supporting my team’s achievement of ranking #1 in quality across DDD branches during peak phase.

As a Quality Assurance Associate and ongoing independent subcontractor, I’ve maintained high QA accuracy across multiple AI training projects while staying remote-first and self-managed. I’m available for remote contract work and thrive in deadline-driven environments where consistent throughput and structured communication matter.