koros benedict

I’m an NLP Data Annotator and AI Trainer focused on hands-on, current experience labeling and validating AI training data for large language models. At Taskap.io, I tag named entities, dates, and domain-specific terminology, and I evaluate LLM-generated responses for accuracy and safety.

My workflow emphasizes instruction following, attention to detail, and structured quality guidelines. I also classify content by sentiment and intent, and I work fluently across English and Kiswahili to support multilingual alignment.

Before moving deeper into NLP data work, I built a strong technical foundation through computer science, structured data management, and IT systems. As an ICT Graduate Trainee (ICTA), I supported software testing, deployment, and technical documentation while managing validated datasets for the Assisted Digital Literacy Program.

I’ve also led hands-on, security- and operations-focused support through the NIIMS project, where I supervised biometric data registration and verification with secure data transmission and backups. In IT Support roles, I handled troubleshooting, permissions, system testing, and reporting—bringing a practical, quality-first mindset to every annotation task.