Treasure

Key Skills

Software

Appen

Lionbridge

Mercor

CVAT

Data Annotation Tech

Clickworker

Mindrift

Micro1

Scale AI

SuperAnnotate

Toloka

Telus

Top Subject Matter

LLM Evaluation and RLHF (Reinforcement Learning from Human Feedback)

AI Safety and Content Moderation

Audio and Speech Annotation

Top Data Types

Text

Audio

Video

Document

Image

Top Task Types

Classification

Transcription

Action Recognition

RLHF

Function Calling

Freelancer Overview

AI Content Creator (Annotation) | Appen | Remote. Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Appen, Lionbridge, and Mercor. Education includes Bachelor of Science, Lagos State University (2021). AI-training focus includes data types such as Text, Audio, and Video and labeling workflows including Classification, Transcription, and Action Recognition.

Labeling Experience

Digital Data Associate and Annotation Generalist | Freelance | Remote

OtherDocumentFunction Calling

Executed high-volume structured data labeling and multimodal annotation across concurrent text, image, and audio projects while maintaining strict quality standards. Designed edge-case and adversarial scenarios for AI safety testing to expose model weaknesses and stress model boundaries. Managed annotation workflows across projects, ensuring delivery timelines and guideline compliance without sacrificing accuracy. • Labeled structured and multimodal examples with quality controls • Created adversarial/edge-case scenarios for safety testing datasets • Ensured 99%+ accuracy through self/QA workflow discipline • Coordinated multiple concurrent annotation assignments

2024 - Present

AI Safety and Prompt Engineering Specialist (Annotation) | Freelance | Remote

Don't discloseTextRLHF

Performed RLHF preference annotation by ranking model responses based on quality, accuracy, helpfulness, and safety. Designed and deployed systematic prompt testing frameworks to evaluate LLM instruction adherence and ethical alignment across model iterations. Authored structured evaluation reports documenting bias triggers and high-risk behavioral patterns with mitigation strategies. • Ranked prompt-response pairs for RLHF fine-tuning support • Conducted linguistic and sentiment analysis on model outputs • Built adversarial, story-based prompts to surface edge-case failures • Delivered evaluation reports to model development teams

2024 - Present

Audio Model Trainer and Data Annotation Specialist | Mercor | Remote

MercorAudioTranscription

Annotated multilingual audio datasets to train and evaluate ASR and natural language understanding systems. Performed QA on ASR transcription outputs by identifying phonetic inconsistencies, transcription errors, and structural misalignments affecting model accuracy. Used precision phonetic annotation to maintain consistency across high-volume audio batches. • Verbatim/intelligent transcription support for accented and noisy audio • Phonetic annotation and pronunciation labeling across languages • Inter-annotator agreement calibration for edge-case handling • Quality assurance to prevent label drift and structural misalignment

2025 - 2025

LLM Trainer and Video Annotation Specialist | Mercor | Remote

MercorVideoAction Recognition

Supported multimodal model development with advanced video labeling for action recognition and contextual scene understanding. Applied structured annotation frameworks to complex or ambiguous video sequences to strengthen dataset coherence and model grounding. Participated in dataset quality reviews by flagging inconsistencies and contributing to guideline refinements to improve inter-annotator agreement. • Labeled actions/activities across temporal video sequences • Performed scene segmentation and temporal event annotation • Flagged annotation inconsistencies during quality reviews • Contributed to scalable annotation guideline refinement

2023 - 2024

Data Annotation Specialist | Louby AI | Remote

TextClassification

Performed comprehensive data annotation and QA across text, structured, and multimodal datasets with ethical governance compliance. Improved annotation workflow efficiency by 18% through identifying bottlenecks and implementing lightweight automation while preventing labeling errors. Conducted self-QA reviews and monitoring to catch label drift and systemic inconsistencies before they propagated into training data. • Self-QA and peer-QA to validate label quality • Applied quality thresholds to reduce model drift in production systems • Flagged systemic inconsistencies and label drift • Maintained platform accuracy benchmarks across long-running projects

2022 - 2024

Show more

Education

Work History