Oyo State, Nigeria
Struggling to train your ML/AI models with messy, inconsistent data? I’m a Data Annotation Expert who delivers high-quality, structured datasets that boost model accuracy and save you time. With 4+ years in tech, I’ve helped companies like Moonvalley and CrowdAi improve their AI performance by 20% through precise annotation. Let’s connect. DM to discuss how I can help your team succeed!
o Recognizing the gap in accessible tech education for people with non-technical backgrounds, I launched G.TecHive, a startup aimed at democratizing entry-level tech training. o My goal was to build a beginner-friendly platform that equips individuals with practical digital and data skills to improve job readiness in today’s tech-driven world. o I developed and delivered tailored training programs focused on the Microsoft Office Suite, Google Workspace, Microsoft 365, Data Annotation for Machine Learning, and Artificial Intelligence. o Positioning G.TecHive as a bridge for non-tech learners into the digital workforce. o Successfully launched the first training cohort with a 100% completion rate and strong participant feedback. Built partnerships with local communities and mentors, establishing G.TecHive as an emerging startup committed to accessible, hands-on tech education.
Produced over 3,000 high-quality captions with a 98% approval rate. This contributed to a significant improvement in model comprehension scores and enhanced dataset value for downstream NLP and CV tasks. The project output was later used in benchmark testing for multimodal AI systems.
Used the company's annotation platform to tag and track objects across frames. Followed strict labeling guidelines, conducted manual QA, and flagged ambiguous cases for cross-checking. Additionally, collaborated with ML engineers to refine class definitions and resolve data ambiguities Achieved over 98% annotation accuracy, contributing to a 15% increase in model precision during validation. Helped streamline annotation workflows, reducing project turnaround time by 20%.
Delivered building annotations with 98% boundary precision, which led to a measurable improvement in the model’s structure detection accuracy and directly supported its deployment in urban planning simulations and geospatial analysis projects.