Airton
Zurich, Switzerland
I will evaluate and improve your LLM outputs with RLHF feedback
(0) Remote 5 months ago
75 $ - Per hour
I help AI teams improve their Large Language Model outputs through structured RLHF (Reinforcement Learning from Human Feedback) evaluation.
With 3+ years of hands-on experience at TaskUs and Translated, I identify weaknesses in model responses, annotate outputs systematically, and develop actionable feedback that measurably improves model quality.
I specialize in Portuguese-language LLM evaluation (PT-PT, PT-BR, PALOP) as well as English, making me especially valuable for teams targeting Lusophone markets. I've built annotation libraries, established quality frameworks, and trained other annotators on best practices.
What I deliver: Structured evaluation reports, annotated datasets, RLHF feedback cycles, and quality improvement recommendations — using annotation tools, quality metric frameworks, and pattern recognition to surface recurring issues.
Available for ongoing evaluation contracts or one-off model audits.
With 3+ years of hands-on experience at TaskUs and Translated, I identify weaknesses in model responses, annotate outputs systematically, and develop actionable feedback that measurably improves model quality.
I specialize in Portuguese-language LLM evaluation (PT-PT, PT-BR, PALOP) as well as English, making me especially valuable for teams targeting Lusophone markets. I've built annotation libraries, established quality frameworks, and trained other annotators on best practices.
What I deliver: Structured evaluation reports, annotated datasets, RLHF feedback cycles, and quality improvement recommendations — using annotation tools, quality metric frameworks, and pattern recognition to surface recurring issues.
Available for ongoing evaluation contracts or one-off model audits.
Please sign in as a customer to give your feedback
