Skip to main content
Godfrey Jonathan

Godfrey Jonathan

AI Trainer • RLHF Evaluator • Web3 Trust & Safety Specialist

Talent
AbujaMember since July 28, 2026
RLHF & Response EvaluationPrompt Engineering & TestingData Annotation & LabelingLLM Training Data CreationContent Moderation & Trust & SafetyRubric Based Quality AssessmentWeb3 & Blockchain Domain KnowledgeMisinformation & Hallucination Detection
·

Hire this person

Send a job offer directly to this candidate

RLHF & Response EvaluationPrompt Engineering & TestingData Annotation & LabelingLLM Training Data CreationContent Moderation & Trust & Safety

About

AI Trainer and RLHF Evaluator with 3+ years of cross domain experience assessing LLM outputs, building annotation workflow, and enforcing trust & safety policies across high volume digital environments. Combines a rare dual expertise in AI evaluation methodology and deep Web3/crypto domain knowledge enabling accurate, context aware review of blockchain, DeFi, and crypto related prompts where most annotators lack fluency. Consistent track record of maintaining quality at scale across RLHF ranking tasks, hallucination flagging, and moderation workflows.

Currently completing a B.Sc. in Computer Science, reinforcing technical foundations in ML and data systems.

Experience

AI Content Evaluator & RLHF Trainer - Outlier AI / Data Annotation / Remotasks - Remote

(2025-01)

  • Reviewed and ranked 15,000+ AI generated responses across RLHF workflows evaluating reasoning quality, accuracy, tone, instruction adherence, and policy compliance.
  • Specialized in crypto and Web3 prompt evaluation, applying domain knowledge to flag factual errors, hallucinations, and misinformation in blockchain/DeFi content.
  • Produced structured prompt response data sets for LLM fine tuning, covering creative writing, technical explanation, and domain specific crypto topics.
  • Applied annotation rubrics and multi-criteria evaluation frameworks to score outputs consistently, maintaining quality across high volume sprints.
  • Flagged unsafe outputs, jail break attempts, and policy violations with detailed evaluation rationales that supported model iteration cycles.

Community Manager & Trust & Safety Lead - Web3 Projects Discord, Telegram, X - Remote

(2024-01)

  • Managed and moderated Discord/Telegram communities handling 500+ daily interactions, enforcing platform rules and real time threat detection (spam, phishing, scam accounts).
  • Hosted and produced X Spaces AMAs, built structured discussion frameworks, managed live moderation, and coordinated speakers for technical sessions.
  • Developed onboarding guides, FAQs, and community content that reduced inbound support queries and improved new member retention.
  • Collaborated directly with founding teams on community growth strategy, reducing misinformation incidents and improving signal to noise quality across channels.

Content Writer & Research Analyst - Web3 & AI Tech - Remote

(2023-01 - 2023-12)

  • Wrote educational articles, project breakdowns, and technical explainers on blockchain, DeFi, NFTs, and AI consistently meeting accuracy and editorial standards.
  • Developed content review frameworks and editorial checklists used across multi-author projects to enforce consistency and factual correctness.
  • Conducted structured fact checking and technical research, building the research methodology skills directly applicable to AI annotation workflows.

Education

B.Sc. - Computer Science - Abubakar Tafawa Balewa University, Bauchi, Nigeria

Skills

AI TrainerRLHF EvaluatorWeb3 Trust & Safety Specialist

Reviews