Adeolu Idowu

Key Skills

Software

Data Annotation Tech

OneForma

Roboflow

Telus

Other

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Image

Text

Video

Top Label Types

Action Recognition

Bounding Box

Segmentation

Classification

Text Summarization

Freelancer Overview

I am an experienced Senior Quality Engineer with a strong background in designing and implementing data-driven testing solutions across regulated industries. My expertise extends to AI training data and data labeling, where I have honed skills in ensuring data accuracy, consistency, and compliance through rigorous quality assurance processes. At TELUS International, as a Data Annotator, I contributed to large-scale AI projects by labeling and annotating datasets for machine learning models, focusing on natural language processing and image recognition tasks, which enhanced model precision and reduced bias. What sets me apart is my ability to bridge QA methodologies with AI data pipelines, leveraging tools like Python and SQL to automate validation workflows and improve data integrity.In my role as an Aether Generalist at Outlier, I worked on diverse annotation projects involving multimodal data, such as text, audio, and visual content, supporting AI model training for applications in autonomous systems and conversational AI. This experience complemented my QA engineering background by applying defect tracking and root cause analysis to identify and rectify annotation errors in real-time, resulting in higher-quality training datasets. Previously, at TELUS, I annotated over 10,000 data points for sentiment analysis and entity recognition projects, collaborating with cross-functional teams to meet stringent deadlines and quality benchmarks. These projects have equipped me with a unique blend of technical acumen and practical insights into AI data ecosystems.

Labeling Experience

Aether Generalist

OtherImageBounding BoxSegmentation

Participated in the Aether Generalist project via the Multimango platform, contributing to multimodal AI data improvement through diverse annotation tasks. Primarily focused on image evaluation and labeling, including entity tagging, captioning, feature identification, text-image pairing, and occasional text editing or content assessment.Completed thousands of high-volume tasks with emphasis on accuracy, following evolving guidelines and maintaining strong quality scores via platform feedback and golden tasks. This supported development of robust generative and multimodal AI systems.

Participated in the Aether Generalist project via the Multimango platform, contributing to multimodal AI data improvement through diverse annotation tasks. Primarily focused on image evaluation and labeling, including entity tagging, captioning, feature identification, text-image pairing, and occasional text editing or content assessment.Completed thousands of high-volume tasks with emphasis on accuracy, following evolving guidelines and maintaining strong quality scores via platform feedback and golden tasks. This supported development of robust generative and multimodal AI systems.

2025 - 2025

Data Annotator

TelusVideoAction Recognition

As a data annotator on this video action recognition project, I labeled short video clips to identify and classify human actions and activities (e.g., walking, running, jumping, waving, interacting with objects, or other predefined action classes). Tasks involved temporal annotation, such as marking the start/end frames of specific actions, bounding box tracking for actors/objects across frames, and assigning action labels using TELUS Digital's Ground Truth Studio platform.The project consisted of approximately [e.g., 500–2000] video clips (adjust to your actual size), sourced from diverse scenarios to ensure model robustness. Quality measures included adherence to detailed annotation guidelines, inter-annotator agreement checks, regular quality audits, and achieving high accuracy. This contributed to training computer vision models for improved action detection in AI applications

As a data annotator on this video action recognition project, I labeled short video clips to identify and classify human actions and activities (e.g., walking, running, jumping, waving, interacting with objects, or other predefined action classes). Tasks involved temporal annotation, such as marking the start/end frames of specific actions, bounding box tracking for actors/objects across frames, and assigning action labels using TELUS Digital's Ground Truth Studio platform.The project consisted of approximately [e.g., 500–2000] video clips (adjust to your actual size), sourced from diverse scenarios to ensure model robustness. Quality measures included adherence to detailed annotation guidelines, inter-annotator agreement checks, regular quality audits, and achieving high accuracy. This contributed to training computer vision models for improved action detection in AI applications

2025 - 2025

Education

Work History