Daniel Ogarah

Key Skills

Software

CVAT

Label Studio

Prodigy

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Task Types

Classification

Object Detection

Text Generation

Text Summarization

Freelancer Overview

As an AI Data Trainer with a background in computer science and cybersecurity, I’ve had the opportunity to work on a variety of projects focused on large language model (LLM) evaluation, prompt engineering, and data annotation across multiple formats—including text, images, and video. I’m particularly skilled at delivering accurate, structured labeling that helps AI systems learn more effectively and responsibly. I've contributed to fine-tuning LLMs by writing and evaluating prompts, rating responses, and ensuring output quality aligned with safety and performance standards. My experience also includes hands-on work with image annotation tasks like classification and bounding boxes, using tools like Label Studio, CVAT, and Prodigy. I take pride in being adaptable, detail-oriented, and dependable—especially when it comes to maintaining data quality and following ethical AI practices.

Labeling Experience

Educational Data Classification

Label StudioTextClassificationQuestion Answering

Contributed to the classification and annotation of educational content, including math problems, instructional text, and student queries, to improve AI-powered tutoring systems. Tasks included categorizing questions by topic and difficulty level, annotating answers for accuracy, and summarizing instructional content to support adaptive learning models. The project handled thousands of text entries and adhered to strict quality assurance protocols to maintain annotation consistency and reliability.

Contributed to the classification and annotation of educational content, including math problems, instructional text, and student queries, to improve AI-powered tutoring systems. Tasks included categorizing questions by topic and difficulty level, annotating answers for accuracy, and summarizing instructional content to support adaptive learning models. The project handled thousands of text entries and adhered to strict quality assurance protocols to maintain annotation consistency and reliability.

2024

Image Annotation for Cybersecurity Training Materials

CVATImageBounding BoxClassification

Worked on annotating cybersecurity-related images to support AI models designed for identifying phishing attempts, malware, and scam indicators in visual formats. Tasks involved drawing bounding boxes around suspicious elements, classifying image content by threat type, and ensuring high-quality, consistent annotations. Maintained strict adherence to data privacy and ethical standards. The project included thousands of images and rigorous quality checks to ensure labeling accuracy and reliability for downstream AI model training.

Worked on annotating cybersecurity-related images to support AI models designed for identifying phishing attempts, malware, and scam indicators in visual formats. Tasks involved drawing bounding boxes around suspicious elements, classifying image content by threat type, and ensuring high-quality, consistent annotations. Maintained strict adherence to data privacy and ethical standards. The project included thousands of images and rigorous quality checks to ensure labeling accuracy and reliability for downstream AI model training.

2023 - 2023

LLM Prompt & Response Evaluation

ProdigyTextClassificationQuestion Answering

Evaluated outputs generated by large language models for tasks such as classification, summarization, instruction following, and question answering in English. Rated model responses on multiple criteria including correctness, tone, relevance, and factual consistency. Assisted in fine-tuning and improving model behavior by providing detailed feedback and annotations, supporting the development of more accurate and responsible AI systems.

Evaluated outputs generated by large language models for tasks such as classification, summarization, instruction following, and question answering in English. Rated model responses on multiple criteria including correctness, tone, relevance, and factual consistency. Assisted in fine-tuning and improving model behavior by providing detailed feedback and annotations, supporting the development of more accurate and responsible AI systems.

2022 - 2023

Education

Work History