Udoh Joy

Key Skills

Software

Remotasks

Data Annotation Tech

Other

Top Subject Matter

General Knowledge

Nigerian & West African Culture

Stem Domain Expertise

Top Data Types

Video

Text

Image

Document

Top Task Types

Data Collection

Text Generation

Classification

Freelancer Overview

Freelance AI Data Specialist (Video Data Collection & Annotation). Brings 9+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Remotasks, Tarmack, and Data Annotation Tech. Education includes Bachelor of Science, Miva (2022). AI-training focus includes data types such as Video and Text and labeling workflows including Data Collection, Text Generation, and Evaluation.

Labeling Experience

Nigerian Context Dataset Initiative (Personal Project)

OtherVideoData Collection

I independently recorded and annotated 75+ videos and images representing authentic Nigerian everyday environments for a model diversity initiative. The project emphasized collection of regional data types to address gaps in Western-trained CV/LLM systems. Key outputs included detailed annotation and identification of 30+ model failure cases. • Collected diverse contextual video/image data for dataset enrichment. • Annotated spatial and action-focused metadata to standard schemas. • Documented model failure modes unique to West African scenes. • Enhanced dataset value for LLM and CV research applications.

I independently recorded and annotated 75+ videos and images representing authentic Nigerian everyday environments for a model diversity initiative. The project emphasized collection of regional data types to address gaps in Western-trained CV/LLM systems. Key outputs included detailed annotation and identification of 30+ model failure cases. • Collected diverse contextual video/image data for dataset enrichment. • Annotated spatial and action-focused metadata to standard schemas. • Documented model failure modes unique to West African scenes. • Enhanced dataset value for LLM and CV research applications.

2025 - Present

Freelance AI Output Evaluator & RLHF Annotator

Data Annotation TechText

I evaluated and rated over 1,500 LLM-generated outputs for helpfulness, harmlessness, truthfulness, and style compliance. I contributed to RLHF and model evaluation tasks across multiple domains with a focus on English language accuracy. My role frequently involved rewriting and enhancing model responses to meet C2 standards. • Scored responses for safety and accuracy, flagging harmful outputs. • Rewrote and edited LLM outputs to improve clarity and specificity. • Applied red teaming and bias detection strategies on edge case prompts. • Supported model improvement through iterative feedback and team QA.

I evaluated and rated over 1,500 LLM-generated outputs for helpfulness, harmlessness, truthfulness, and style compliance. I contributed to RLHF and model evaluation tasks across multiple domains with a focus on English language accuracy. My role frequently involved rewriting and enhancing model responses to meet C2 standards. • Scored responses for safety and accuracy, flagging harmful outputs. • Rewrote and edited LLM outputs to improve clarity and specificity. • Applied red teaming and bias detection strategies on edge case prompts. • Supported model improvement through iterative feedback and team QA.

2023 - Present

Freelance AI Data Specialist (Text Annotation & Edge Case Generation)

TextText Generation

I crafted and fact-checked over 600 free-form text responses designed to train large language models in reasoning, logic, and domain expertise. Work involved strict adherence to C2 English standards and comprehensive QA processes. I generated nuanced, context-aware prompts and answers for both general knowledge and STEM domains. • Wrote and reviewed LLM training data for enhanced logical and cultural depth. • Consistently achieved a 98.5% QA pass rate on text-based response projects. • Produced 200+ culturally-relevant edge cases using local context. • Implemented fast QA feedback, improving team revision turnaround times.

I crafted and fact-checked over 600 free-form text responses designed to train large language models in reasoning, logic, and domain expertise. Work involved strict adherence to C2 English standards and comprehensive QA processes. I generated nuanced, context-aware prompts and answers for both general knowledge and STEM domains. • Wrote and reviewed LLM training data for enhanced logical and cultural depth. • Consistently achieved a 98.5% QA pass rate on text-based response projects. • Produced 200+ culturally-relevant edge cases using local context. • Implemented fast QA feedback, improving team revision turnaround times.

2023 - Present

Freelance AI Data Specialist (Video Data Collection & Annotation)

RemotasksVideoData Collection

As a freelance AI Data Specialist, I performed multimodal video data collection and annotation across various projects. I ensured video data quality by adhering to strict QC standards and technical requirements for mobile ultra-wide cameras. My responsibilities included creating video datasets, documenting technical compliance, and maintaining high approval rates. • Captured and organized video data using advanced smartphone videography standards. • Maintained a 97.2% first-pass quality control rate over 400+ video recording and SFT tasks. • Authored detailed metadata for 300+ videos/images using JSON schemas. • Diagnosed and documented device compatibility securing platform approvals.

As a freelance AI Data Specialist, I performed multimodal video data collection and annotation across various projects. I ensured video data quality by adhering to strict QC standards and technical requirements for mobile ultra-wide cameras. My responsibilities included creating video datasets, documenting technical compliance, and maintaining high approval rates. • Captured and organized video data using advanced smartphone videography standards. • Maintained a 97.2% first-pass quality control rate over 400+ video recording and SFT tasks. • Authored detailed metadata for 300+ videos/images using JSON schemas. • Diagnosed and documented device compatibility securing platform approvals.

2023 - Present

AI Content Quality Lead

RemotasksVideoClassification

I led quality assessment and review of video and text annotations created by junior contractors. My responsibilities included guideline adherence checks, feedback provision, and maintaining high QC rates for the team. I also developed onboarding and training documentation adopted platform-wide. • Reviewed and improved accuracy of 1,000+ video and text annotations. • Implemented a feedback system raising team quality rates from 89% to 96%. • Developed standard onboarding materials for annotation processes. • Oversaw project-wide labeling accuracy and documentation.

I led quality assessment and review of video and text annotations created by junior contractors. My responsibilities included guideline adherence checks, feedback provision, and maintaining high QC rates for the team. I also developed onboarding and training documentation adopted platform-wide. • Reviewed and improved accuracy of 1,000+ video and text annotations. • Implemented a feedback system raising team quality rates from 89% to 96%. • Developed standard onboarding materials for annotation processes. • Oversaw project-wide labeling accuracy and documentation.

2023 - 2024

Education

Work History