Uzoamaka Onyewuchi

Key Skills

Software

Lionbridge

Snorkel AI

Other

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Audio

Top Label Types

Classification

RLHF

Evaluation Rating

Audio Recording

Transcription

Freelancer Overview

I am a detail-oriented data annotator and AI evaluator with hands-on experience labeling and reviewing text, image, video, and audio data for various AI projects. My background in psychology allows me to analyze behavior, interpret patterns, and enhance dataset quality and reliability. I am skilled in identifying errors, enriching datasets with structured metadata, and providing clear, structured feedback to support AI model training and evaluation. I am proficient with tools such as Ubuntu Linux, open-source applications like LibreOffice and GIMP, and basic JSON editing. I consistently follow detailed guidelines, work independently, and maintain high accuracy under tight deadlines, ensuring that data is accurate, consistent, and ready for use in AI applications.

Labeling Experience

AI Response evaluation & RLHF Data Annotation Specialist

Snorkel AITextRLHFEvaluation Rating

Worked on large-scale text annotation and AI response evaluation projects supporting large language model training. Assessed model-generated outputs based on structured rubrics, including accuracy, relevance, safety, reasoning quality, and policy compliance. Tasks involved comparing multiple responses, identifying hallucinations, flagging unsafe or misleading content, and providing structured ratings to improve model alignment. Operated within Snorkel AI's guideline-driven annotations, ensuring consistency across high-volume tasks while adapting to evolving project instructions. Maintained strong quality standards through calibration exercises, feedback loops, and strict adherence to evaluation criteria to support reliable model fine-tuning and performance optimization.

Worked on large-scale text annotation and AI response evaluation projects supporting large language model training. Assessed model-generated outputs based on structured rubrics, including accuracy, relevance, safety, reasoning quality, and policy compliance. Tasks involved comparing multiple responses, identifying hallucinations, flagging unsafe or misleading content, and providing structured ratings to improve model alignment. Operated within Snorkel AI's guideline-driven annotations, ensuring consistency across high-volume tasks while adapting to evolving project instructions. Maintained strong quality standards through calibration exercises, feedback loops, and strict adherence to evaluation criteria to support reliable model fine-tuning and performance optimization.

2025

Audio Transcription & Annotation for AI Training

OtherAudioAudio RecordingTranscription

Performed audio transcription and annotation tasks for Babel Audio AI training datasets, focusing on accurately converting speech to text while following strict linguistic and formatting guidelines. Reviewed audio clips for clarity, speaker intent, and contextual meaning, ensuring high transcription accuracy and consistency across diverse accents and recording conditions. Also handled quality checks, corrections, and guideline adherence to maintain dataset reliability for speech and language model development.

Performed audio transcription and annotation tasks for Babel Audio AI training datasets, focusing on accurately converting speech to text while following strict linguistic and formatting guidelines. Reviewed audio clips for clarity, speaker intent, and contextual meaning, ensuring high transcription accuracy and consistency across diverse accents and recording conditions. Also handled quality checks, corrections, and guideline adherence to maintain dataset reliability for speech and language model development.

2024

AI Data Labeling & Model Evaluation Specialist

LionbridgeImageClassification

Contributed to an image classification project supporting large language model training and evaluation. The scope involved reviewing and categorizing image datasets according to predefined taxonomy guidelines to improve dataset organization and model alignment. Tasks included assigning appropriate labels to images based on visual content, identifying edge cases, and flagging ambiguous or policy-sensitive material for secondary review. Worked within Lionbridge’s annotation platform, adhering strictly to detailed labeling instructions and quality benchmarks. The project required high attention to detail, consistency across large data batches, and regular calibration against updated guidelines. Maintained strong accuracy standards through self-review and quality feedback cycles, ensuring reliable dataset integrity for downstream AI model development.

Contributed to an image classification project supporting large language model training and evaluation. The scope involved reviewing and categorizing image datasets according to predefined taxonomy guidelines to improve dataset organization and model alignment. Tasks included assigning appropriate labels to images based on visual content, identifying edge cases, and flagging ambiguous or policy-sensitive material for secondary review. Worked within Lionbridge’s annotation platform, adhering strictly to detailed labeling instructions and quality benchmarks. The project required high attention to detail, consistency across large data batches, and regular calibration against updated guidelines. Maintained strong accuracy standards through self-review and quality feedback cycles, ensuring reliable dataset integrity for downstream AI model development.

2024

Education

Work History