Oluwaseun Adeleye

Key Skills

Software

Scale AI

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Label Types

Action Recognition

Audio Recording

Bounding Box

Classification

Entity Ner Classification

Evaluation Rating

Prompt Response Writing SFT

Question Answering

Text Generation

Text Summarization

Transcription

Freelancer Overview

I am a detail-oriented AI data annotator and quality analyst with over five years of experience in text annotation, search relevance evaluation, policy-based content review, and human-in-the-loop quality assurance. My background includes labeling and validating large datasets for AI model training and evaluation, with a strong focus on natural language processing, search engine relevance, and content moderation. I am skilled in applying complex guidelines, performing entity recognition, error categorization, and ground-truth data creation using tools such as Power BI, R, Stata, and SQL. My experience spans evaluating AI-generated outputs, moderating web content, and ensuring linguistic and policy accuracy for production AI systems. I thrive in fast-paced, remote environments, consistently delivering high-quality, accurate annotations and constructive feedback to improve model performance and data quality.

Labeling Experience

AI Image & Multimodal Output Evaluation (Generative AI)

Scale AIImageBounding Box

Performed image selection and post-processing using brush-based masking to remove undesired objects or artifacts from AI-generated images. Guided controlled regeneration of masked regions to produce visually consistent replacements, ensuring structural continuity, accurate context blending, and adherence to prompt constraints. Evaluated regenerated outputs against quality rubrics and rejected outputs with artifacts, inconsistencies, or prompt drift.

Performed image selection and post-processing using brush-based masking to remove undesired objects or artifacts from AI-generated images. Guided controlled regeneration of masked regions to produce visually consistent replacements, ensuring structural continuity, accurate context blending, and adherence to prompt constraints. Evaluated regenerated outputs against quality rubrics and rejected outputs with artifacts, inconsistencies, or prompt drift.

2025 - 2025

Education

Work History