For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Magdalena O’Grady

Magdalena O’Grady

Adversarial Testing & Annotation Projects (Remote, Freelance)

USA flag
New York City, Usa
$35.00/hrIntermediateInternal Proprietary ToolingOther

Key Skills

Software

Internal/Proprietary Tooling
Other

Top Subject Matter

Large Language Models Security and Safety
Large Language Model Vulnerability Assessment
AI Model Safety and Response Evaluation

Top Data Types

TextText
DocumentDocument
ImageImage

Top Task Types

Red Teaming
Evaluation Rating
RLHF

Freelancer Overview

Adversarial Testing & Annotation Projects (Remote, Freelance). Core strengths include Internal and Proprietary Tooling. AI-training focus includes data types such as Text and labeling workflows including Red Teaming, Evaluation, and Rating.

IntermediatePolishEnglish

Labeling Experience

Alignerr.com: Active AI Trainer/Annotator/Judge

TextRed Teaming
Actively judged and provided feedback in the context of AI training and response evaluation. Participated in rating and reviewing model outputs across various projects. Delivered detailed written feedback for improvement. • Response evaluation and judgment • Detailed feedback on task output • Consistent quality in task ratings • Participation in multiple active projects

Actively judged and provided feedback in the context of AI training and response evaluation. Participated in rating and reviewing model outputs across various projects. Delivered detailed written feedback for improvement. • Response evaluation and judgment • Detailed feedback on task output • Consistent quality in task ratings • Participation in multiple active projects

2025 - Present

Ongoing Adversarial & Red Teaming Work (Remote, Contract)

TextRed Teaming
Conducted adversarial testing by prompting injections and exposing vulnerabilities in language models. Produced annotated exploit datasets and wrote risk analysis reports. Handled tasks in English with additional Polish developer support. • Structured and reproducible adversarial annotation • Vulnerability flagging and dataset annotation • Risk report generation • Multilingual dataset evaluation

Conducted adversarial testing by prompting injections and exposing vulnerabilities in language models. Produced annotated exploit datasets and wrote risk analysis reports. Handled tasks in English with additional Polish developer support. • Structured and reproducible adversarial annotation • Vulnerability flagging and dataset annotation • Risk report generation • Multilingual dataset evaluation

2024 - Present

DataAnnotation: Active AI Trainer/Annotator

TextRed Teaming
Worked as an AI Trainer and Annotator for DataAnnotation, reviewing and annotating AI-generated responses. Focused on quality, safety, and usefulness of outputs. Delivered active work on annotation and feedback tasks. • Annotated and rated AI model outputs • Provided feedback for model training • Ensured response quality and safety • Collaborated on ongoing annotation projects

Worked as an AI Trainer and Annotator for DataAnnotation, reviewing and annotating AI-generated responses. Focused on quality, safety, and usefulness of outputs. Delivered active work on annotation and feedback tasks. • Annotated and rated AI model outputs • Provided feedback for model training • Ensured response quality and safety • Collaborated on ongoing annotation projects

2023 - Present

Independent AI Trainer & Red Teamer Micro One, Freelance Remote

TextRed Teaming
Worked on flexible AI training and red team projects centered around adversarial input annotation and edge case evaluation. Engaged in daily probing of model outputs for failures and safety issues. Built core English context and provided annotation-based feedback for model improvement. • Annotated adversarial and bias test data • Performed real-time stress tests on AI generators • Collaborated on red-teaming of ML tools • Provided safety and bias evaluation annotations

Worked on flexible AI training and red team projects centered around adversarial input annotation and edge case evaluation. Engaged in daily probing of model outputs for failures and safety issues. Built core English context and provided annotation-based feedback for model improvement. • Annotated adversarial and bias test data • Performed real-time stress tests on AI generators • Collaborated on red-teaming of ML tools • Provided safety and bias evaluation annotations

2022 - Present

Adversarial Testing & Annotation Projects (Remote, Freelance)

TextRed Teaming
Manual adversarial testing of large language models focused on testing vulnerabilities and stress points. Crafted and annotated datasets, provided detailed notes, and delivered feedback for improvements. Involved scoring on benchmarks and supporting English and Polish language tasks. • Conducted jailbreak and edge case prompt creation • Scored and annotated adversarial test datasets • Provided vulnerability and improvement documentation • Bilingual support (English/Polish) for global checks

Manual adversarial testing of large language models focused on testing vulnerabilities and stress points. Crafted and annotated datasets, provided detailed notes, and delivered feedback for improvements. Involved scoring on benchmarks and supporting English and Polish language tasks. • Conducted jailbreak and edge case prompt creation • Scored and annotated adversarial test datasets • Provided vulnerability and improvement documentation • Bilingual support (English/Polish) for global checks

2023 - 2025

Education

U

University of Warsaw

Bachelor, Human Resource

Bachelor
1995 - 2000

Work History

S

Solar and Roofing

Operations Director

New York
2025 - Present
B

BLDG Management

Senior Property Manager

New York
2021 - 2025