Sunday Akinbamido

Key Skills

Software

Clickworker

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Image

Text

Video

Top Label Types

Action Recognition

Audio Recording

Classification

Computer Programming Coding

Emotion Recognition

Entity Ner Classification

Evaluation Rating

Function Calling

Freelancer Overview

I have worked as a freelance AI coding trainer and on data annotation for video, audio, and text across large datasets, contributing to the preparation and refinement of high-quality AI training data. My responsibilities included labeling, validating, and reviewing structured and unstructured data while ensuring accuracy, consistency, and adherence to annotation guidelines. In addition, I bring a strong technical background from building API-driven platforms, particularly in fintech, where I work extensively with structured data, JSON, and large-scale datasets. This combination of annotation experience and software engineering expertise sets me apart, enabling me to understand how training data is generated, validated, and ultimately used to improve AI and machine learning models.

Labeling Experience

Image Tagging

ClickworkerImageBounding BoxPolygon

Worked on image tagging and object detection, accurately identifying and labeling objects using bounding boxes in accordance with annotation guidelines. The role also involved researching specific items or named individuals via Google Search, capturing relevant screenshots, and, where required, submitting verified source page URLs. This work demanded strong attention to detail, consistency in labeling, and strict adherence to quality standards to ensure reliable datasets for computer vision and AI model training.

Worked on image tagging and object detection, accurately identifying and labeling objects using bounding boxes in accordance with annotation guidelines. The role also involved researching specific items or named individuals via Google Search, capturing relevant screenshots, and, where required, submitting verified source page URLs. This work demanded strong attention to detail, consistency in labeling, and strict adherence to quality standards to ensure reliable datasets for computer vision and AI model training.

2025 - 2025

Voice Rating

ClickworkerAudioClassificationEvaluation Rating

Worked on audio classification and evaluation, rating AI-generated audio outputs to determine which sounded more natural and human-like. The task also involved identifying British accents, distinguishing them from other accents, and classifying voice gender. This required careful listening to tone, pronunciation, rhythm, and clarity, ensuring accurate labeling and consistent ratings in line with defined quality and evaluation guidelines.

Worked on audio classification and evaluation, rating AI-generated audio outputs to determine which sounded more natural and human-like. The task also involved identifying British accents, distinguishing them from other accents, and classifying voice gender. This required careful listening to tone, pronunciation, rhythm, and clarity, ensuring accurate labeling and consistent ratings in line with defined quality and evaluation guidelines.

2025 - 2025

Data Annotation

Scale AIVideoClassificationEvaluation Rating

The work involved video data annotation, including tagging products and identifying people appearing in video content. As part of the task, I conducted internet-based image research, using tools such as Google Image Search to find visually similar product images or reference images of individuals extracted from video frames. These images were then accurately matched and attached to the corresponding annotations to enrich the dataset, improve labeling accuracy, and support downstream AI model training and validation.

The work involved video data annotation, including tagging products and identifying people appearing in video content. As part of the task, I conducted internet-based image research, using tools such as Google Image Search to find visually similar product images or reference images of individuals extracted from video frames. These images were then accurately matched and attached to the corresponding annotations to enrich the dataset, improve labeling accuracy, and support downstream AI model training and validation.

2025 - 2025

Data Annotation

ClickworkerVideoEmotion Recognition

The task involved carefully watching video recordings and accurately tagging the emotions displayed by individuals, such as happy, sad, neutral, and smiling. This required close attention to facial expressions, body language, and subtle emotional cues to ensure consistent and high-quality annotations in line with established labeling guidelines.

The task involved carefully watching video recordings and accurately tagging the emotions displayed by individuals, such as happy, sad, neutral, and smiling. This required close attention to facial expressions, body language, and subtle emotional cues to ensure consistent and high-quality annotations in line with established labeling guidelines.

2025 - 2025

AI Coding Trainer

Scale AIComputer Code ProgrammingEvaluation Rating

Worked as a freelance AI trainer, where I designed and presented effective prompts, reviewed and evaluated AI-generated code responses, and compared multiple outputs (e.g., Response A vs. Response B). I provided well-structured justifications for why one response was better, worse, or equally suitable, strictly following established evaluation guidelines. This role required assessing code correctness, efficiency, readability, and alignment with the prompt, as well as assigning accurate ratings to ensure consistent, high-quality feedback for model improvement.

Worked as a freelance AI trainer, where I designed and presented effective prompts, reviewed and evaluated AI-generated code responses, and compared multiple outputs (e.g., Response A vs. Response B). I provided well-structured justifications for why one response was better, worse, or equally suitable, strictly following established evaluation guidelines. This role required assessing code correctness, efficiency, readability, and alignment with the prompt, as well as assigning accurate ratings to ensure consistent, high-quality feedback for model improvement.

2024 - 2024

Education

Work History