Ozisiaka Arthur

Key Skills

Software

Appen

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Video

Top Label Types

Text Generation

Object Detection

Action Recognition

Text Summarization

Translation Localization

Freelancer Overview

I am a detail-oriented data annotator with hands-on experience in labeling and annotating image, text, video, and audio datasets for AI and machine learning projects. My expertise includes using industry-leading tools like Labelbox, CVAT, Supervisely, and Scale AI platforms, as well as specialized interfaces for tasks such as bounding boxes, semantic segmentation, keypoint annotation, OCR correction, and audio transcription. I am highly skilled at following complex annotation guidelines, maintaining strict data quality standards, and performing thorough QA and consistency checks to ensure the integrity of training data. My background in data analysis, combined with proficiency in Excel, SQL, Power BI, and Python, allows me to support data-driven decision-making and streamline annotation workflows. I am committed to delivering accurate, high-quality labeled data that enhances model performance and am quick to adapt to new tools and project requirements.

Labeling Experience

Data Annotation

AppenTextText GenerationObject Detection

Evaluating AI prompts and their generated responses to ensure they meet defined quality and safety standards. The task requires carefully reading the prompt, analyzing the AI’s response, and assessing whether it is relevant, accurate, clear, complete, and aligned with project guidelines. Reviewers check for issues such as hallucinations, logical errors, bias, harmful or unsafe content, and policy violations. They then assign ratings based on structured scoring criteria and provide concise justifications for their evaluations. The overall goal is to improve AI model performance by ensuring outputs are helpful, coherent, factually correct, and compliant with annotation guidelines.

Evaluating AI prompts and their generated responses to ensure they meet defined quality and safety standards. The task requires carefully reading the prompt, analyzing the AI’s response, and assessing whether it is relevant, accurate, clear, complete, and aligned with project guidelines. Reviewers check for issues such as hallucinations, logical errors, bias, harmful or unsafe content, and policy violations. They then assign ratings based on structured scoring criteria and provide concise justifications for their evaluations. The overall goal is to improve AI model performance by ensuring outputs are helpful, coherent, factually correct, and compliant with annotation guidelines.

2024 - 2025

Education

Work History