Precious O.

Key Skills

Software

Labelbox

iMerit

Mercor

Other

Top Subject Matter

Video action and object interaction annotation for AI training datasets.

LLM output evaluation and feedback for AI training.

Legal Services & Contract Review

Top Data Types

Text

Image

Audio

Top Task Types

Evaluation/Rating

Bounding Box

Classification

Segmentation

Entity (NER) Classification

RLHF

Freelancer Overview

Experienced Detail-oriented AI Training and data annotation specialist with a strong background in LLM evaluation, image review, and quality assurance for AI model development. Skilled in handling text, image, and audio datasets with high accuracy while following complex annotation guidelines and maintaining consistency across large-scale projects. Adept at strong attention to detail, error detection, timestamp alignment, and quality output. Worked on AI training and Annotation projects on platforms such as Atlas Capture, Turing AI, and imerit, contributing to tasks such as atomic action labelling, AI response evaluation, Prompt response & review, Visual QA and image classification. I bring a combination of technical annotation expertise, guideline compliance, and quality-focused review processes that directly support reliable AI model performance. Education includes a Bachelor of Science, Prime Royal University (2027).

Labeling Experience

AI Training Contributor | Remote (Turing)

Text

Evaluated AI-generated outputs for accuracy, reasoning quality, and instruction adherence using structured evaluation guidelines. Assessed responses for logical consistency, factual correctness, and clarity, then provided feedback to improve model-generated content. Contributed to prompt-response analysis and reasoning validation tasks to support AI training. • Accuracy and instruction adherence evaluation for model responses. • Logical consistency and factual correctness assessment. • Apply scoring/rating guidelines to rate and improve outputs. • Provide detailed feedback to enhance training data and performance.

2026 - Present

AI Data Annotator | Remote Contract (Atlas Capture)

VideoAction Recognition

Annotated video datasets using structured labeling frameworks, including action labeling and object interaction tagging. Validated timestamp alignment so annotations precisely correspond to video frames and reviewed outputs for guideline compliance. Documented step-by-step reasoning traces to support training LLMs to reason reliably. • Action labeling and object interaction tagging for video content. • Timestamp alignment checking between annotations and frames. • Correction of inconsistencies based on detailed guidelines and taxonomies. • Quality assurance to maintain dataset accuracy and consistency.

2024 - Present

Education

Work History