Agnes E.

Key Skills

Software

Don't disclose

Appen

Top Subject Matter

Multimodal AI datasets and RLHF evaluation

Content evaluation

multimodal annotation

Top Data Types

Text

Image

Video

Document

Top Task Types

RLHF

Freelancer Overview

TURING — Visual Task Labeling Specialist (Freelance / Remote). Core strengths include Don't disclose and Appen. Education includes Master of Engineering, Nigerian Defence Academy (2025) and Bachelor of Engineering, Federal University of Agriculture, Makurdi (2019). AI-training focus includes data types such as Text, Image, and Video and labeling workflows including RLHF, Evaluation, and Rating.

Labeling Experience

TURING — Visual Task Labeling Specialist (Freelance / Remote)

Don't discloseTextRLHF

Executed visual task labeling across text, image, and video datasets for machine learning model training. Performed RLHF-related tasks by evaluating and ranking model responses to improve truthfulness, helpfulness, and safety. Conducted quality checks to ensure labeled outputs met complex project requirements. • Labeled multimodal data (text, images, video) • Identified annotation errors and visual artifacts • Achieved quality scores of 98% or higher on RLHF • Provided feedback to support continuous AI/data quality improvement

2026 - 2026

APPEN — AI Content Evaluator & Annotator (Freelance / Remote)

AppenTextRLHF

Produced high-quality labels for text, image, and video datasets used to train machine learning models. Evaluated AI-generated responses for factual accuracy, logical flow, and brand-safe tone and provided detailed refinement feedback. Completed RLHF tasks, maintaining consistently high quality scores. • Performed multimodal data labeling (text, images, video) • Conducted evaluation/rating of AI responses • Executed RLHF tasks with 98%+ quality scores • Participated in guideline refinement and edge-case identification

2023 - 2025

RENT TO OWN — Data Researcher & Quality Analyst (Remote)

Don't discloseDocument

Conducted deep-dive research to verify claims made by generative AI models to support output reliability. Categorized large volumes of unstructured data into defined taxonomies to improve search engine algorithms. Identified and flagged biased or harmful content within training sets to support ethical AI development. • Verified generative AI claims for reliability • Performed taxonomy-based categorization of unstructured data • Audited datasets for bias/harmful content • Met/exceeded throughput targets while maintaining quality control

2021 - 2022

Education

Work History