Gideon Odiah

Key Skills

Software

Micro1

Other

Telus

Appen

Top Subject Matter

Natural Language Processing (NLP)

Llms Domain Expertise

AI model QA

Top Data Types

Text

Image

Document

Top Task Types

Prompt + Response Writing (SFT)

Classification

Fine-tuning

Freelancer Overview

Prompt Writer & QA Data Annotation Specialist. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Micro1, Other, and Telus. Education includes Postgraduate Diploma, Babcock University (2025) and Bachelor of Science, Ekiti State University (2021). AI-training focus includes data types such as Text and labeling workflows including Prompt + Response Writing (SFT), Classification, and Evaluation.

Labeling Experience

Prompt Writer & QA Data Annotation Specialist

Micro1TextPrompt Response Writing SFT

As a Prompt Writer & QA Data Annotation Specialist at Micro1 AI, I designed and optimized prompts for large language models to improve their abilities in summarization, classification, and entity extraction. I evaluated thousands of AI-generated outputs to detect tone, contextual accuracy, and hallucination, directly impacting model reliability. I refined annotation workflows and collaborated with team members to enhance project efficiency. • Created and optimized 500+ LLM prompts for diverse NLP tasks. • Assessed 2,000+ outputs for QA, implementing hallucination detection. • Built reusable prompt templates for efficient project scaling. • Led workflow improvements through structured feedback and A/B testing.

As a Prompt Writer & QA Data Annotation Specialist at Micro1 AI, I designed and optimized prompts for large language models to improve their abilities in summarization, classification, and entity extraction. I evaluated thousands of AI-generated outputs to detect tone, contextual accuracy, and hallucination, directly impacting model reliability. I refined annotation workflows and collaborated with team members to enhance project efficiency. • Created and optimized 500+ LLM prompts for diverse NLP tasks. • Assessed 2,000+ outputs for QA, implementing hallucination detection. • Built reusable prompt templates for efficient project scaling. • Led workflow improvements through structured feedback and A/B testing.

2025 - Present

AI Operations Associate

OtherTextClassification

As an AI Operations Associate at Hugotech.co, I led annotation for multimodal datasets involving both NLP and computer vision. I developed and refined prompt libraries and task-specific instructions to support model instruction-following and moderation improvements. My role included auditing outputs for relevance, bias, and contextual accuracy for scalable LLM optimization. • Managed annotation for 50K+ samples across NLP and computer vision tasks. • Developed 80+ prompts for LLM fine-tuning and moderation. • Audited 5,000+ AI outputs for bias and accuracy. • Implemented company-wide reusable prompt libraries.

As an AI Operations Associate at Hugotech.co, I led annotation for multimodal datasets involving both NLP and computer vision. I developed and refined prompt libraries and task-specific instructions to support model instruction-following and moderation improvements. My role included auditing outputs for relevance, bias, and contextual accuracy for scalable LLM optimization. • Managed annotation for 50K+ samples across NLP and computer vision tasks. • Developed 80+ prompts for LLM fine-tuning and moderation. • Audited 5,000+ AI outputs for bias and accuracy. • Implemented company-wide reusable prompt libraries.

2024 - 2025

Data Annotator & AI Content Reviewer

TelusText

At Telus International, I reviewed and tagged over ten thousand NLP model outputs for tasks such as sentiment analysis and classification. My duties included identifying and correcting biased or hallucinated outputs to increase dataset reliability for LLM training. I provided structured evaluation feedback to inform iterative improvements to the models. • Conducted quality assurance on 10K+ NLP outputs. • Flagged more than 1,200 biased or hallucinated AI generations. • Delivered consistent evaluation metrics for model development. • Enhanced ethical compliance and dataset quality.

At Telus International, I reviewed and tagged over ten thousand NLP model outputs for tasks such as sentiment analysis and classification. My duties included identifying and correcting biased or hallucinated outputs to increase dataset reliability for LLM training. I provided structured evaluation feedback to inform iterative improvements to the models. • Conducted quality assurance on 10K+ NLP outputs. • Flagged more than 1,200 biased or hallucinated AI generations. • Delivered consistent evaluation metrics for model development. • Enhanced ethical compliance and dataset quality.

2023 - 2024

AI Data Contributor

AppenTextFine Tuning

As an AI Data Contributor for Appen / Crowdgen, I labeled and verified multimodal data samples for NLP and computer vision model training. My work included assisting in LLM prompt tuning and reducing errors in annotation workflows. This experience provided hands-on exposure to both manual labeling and prompt optimization for large datasets. • Labeled and verified 20K+ multimodal samples across text and vision tasks. • Supported prompt tuning for LLM accuracy improvements. • Reduced annotation errors by 17% through workflow optimization. • Collaborated remotely with international teams for dataset quality.

As an AI Data Contributor for Appen / Crowdgen, I labeled and verified multimodal data samples for NLP and computer vision model training. My work included assisting in LLM prompt tuning and reducing errors in annotation workflows. This experience provided hands-on exposure to both manual labeling and prompt optimization for large datasets. • Labeled and verified 20K+ multimodal samples across text and vision tasks. • Supported prompt tuning for LLM accuracy improvements. • Reduced annotation errors by 17% through workflow optimization. • Collaborated remotely with international teams for dataset quality.

2021 - 2023

Education

Work History