Etinosa O.

Key Skills

Software

Don't disclose

Top Subject Matter

AI safety & Data annotation

trust & safety

adversarial evaluation of conversational models

Top Data Types

Text

Document

Top Task Types

Red Teaming

Freelancer Overview

AI Red-Teaming Contributor (Contract, Remote). Brings 5+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Don't disclose. Education includes Community Manager & Trust and Safety Moderator, Remote (2022). AI-training focus includes data types such as Text and labeling workflows including Red Teaming, Evaluation, and Rating.

Labeling Experience

AI Red-Teaming Contributor (Contract, Remote)

Don't discloseTextRed Teaming

Conduct adversarial testing on conversational AI systems to identify jailbreaks, prompt injections, and misuse scenarios. Probe models for bias, misinformation, harmful behavior, and policy evasion using structured evaluation frameworks. Annotate AI failures by classifying vulnerabilities and flagging systemic risks for remediation. • Created reproducible red-team artifacts including attack cases and written reports. • Produced labeled datasets of AI failures to support safety evaluation and retraining. • Expanded evaluation coverage beyond automated tests through human-led probing. • Documented and packaged findings to enable actionable risk remediation.

2025 - Present

Community Manager & Trust and Safety Moderator (Remote)

Don't discloseText

Moderated and safeguarded high-activity online and Web3 communities by identifying coordinated abuse and harmful user behavior patterns. Stress-tested moderation policies and enforcement tooling to uncover blind spots and edge cases requiring governance updates. Documented recurring risk vectors to support long-term platform safety improvements. • Reviewed user-generated content for abuse, impersonation, scams, disinformation, and social engineering. • Classified and reported risks back to technical and project leadership for corrective action. • Contributed to evaluation of moderation effectiveness through human judgment. • Maintained structured documentation of observed safety failures and recurring threats.

2022 - Present

Education

Work History