Mateo Shija

I’ve spent over two decades delivering dependable enterprise ICT support—network administration, systems infrastructure operations, and IT data quality control—ensuring high availability, performance, and reliability. I built my career on proactively resolving issues, strengthening stability and security, and supporting structured digital data workflows that keep organizations running.

Now I focus on AI data annotation, quality control, and dataset preparation across text, audio, and image modalities. In my current role, I prepare AI-ready multimodal datasets, run NLP annotation (sentiment analysis, intent classification, and NER), perform audio transcription with timestamping and speaker labeling, and complete computer vision labeling with bounding boxes and segmentation—especially for English–Swahili code-switching. I also design annotation guidelines for scalable consistency and support model development inputs (including CNN and VGG16) by validating and correcting data to improve training reliability.