Oluwakayode Tijani

Key Skills

Software

CVAT

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Bounding Box

Freelancer Overview

I am a detail-oriented AI evaluator and data annotator with hands-on experience in evaluating large language model (LLM) outputs for accuracy, safety, coherence, and relevance. My background includes preference ranking (RLHF), prompt writing, text summarization, and thorough error detection—such as identifying hallucinations, bias, and toxicity. I am skilled at following complex guidelines and consistently deliver high-quality results, maintaining accuracy rates of 95–98% in high-volume, repetitive evaluation tasks. I am fluent in both English and Arabic, and I have worked across domains like general knowledge, technology, education, business, and current events. My toolkit includes web-based evaluation platforms, Google Workspace, and spreadsheets, and I bring strong analytical and critical thinking skills to every project.

Labeling Experience

Image Annotation & Object Detection for Autonomous Driving Dataset

CVATImageBounding Box

Contributed high-precision data labeling to support computer vision models in the autonomous driving domain (IT/AI sector). Annotated over 50,000 image frames using CVAT, focusing on bounding box and polygon annotations for key objects including vehicles, pedestrians, cyclists, traffic signs, traffic lights, road lanes, and drivable areas. Strictly adhered to detailed client guidelines, including edge cases like occlusions, low-light conditions, adverse weather, and crowded urban scenes. Achieved consistent high accuracy with >97% Intersection over Union (IoU) in quality reviews and inter-annotator agreement checks. Utilized CVAT features such as automatic interpolation, attribute tagging, task assignment, and review workflows to maintain efficiency and team collaboration. This project enhanced perception systems for safer autonomous vehicle navigation

Contributed high-precision data labeling to support computer vision models in the autonomous driving domain (IT/AI sector). Annotated over 50,000 image frames using CVAT, focusing on bounding box and polygon annotations for key objects including vehicles, pedestrians, cyclists, traffic signs, traffic lights, road lanes, and drivable areas. Strictly adhered to detailed client guidelines, including edge cases like occlusions, low-light conditions, adverse weather, and crowded urban scenes. Achieved consistent high accuracy with >97% Intersection over Union (IoU) in quality reviews and inter-annotator agreement checks. Utilized CVAT features such as automatic interpolation, attribute tagging, task assignment, and review workflows to maintain efficiency and team collaboration. This project enhanced perception systems for safer autonomous vehicle navigation

2024 - 2025

Education

Work History