AI Trainer - RLHF & Text Annotation
I executed high-complexity RLHF tasks including ranking model responses for accuracy, safety, and technical reasoning. My role involved interacting with natural language models and providing human feedback to improve their performance. The outcomes contributed to robust AI systems for organizations such as OpenAI and Anthropic. • Ranked and evaluated language model completions. • Identified and flagged safety and technical errors. • Provided structured feedback for continuous model refinement. • Worked on Remotasks and Toloka for annotation tasks.