AI Training Specialist (Oracle Member) – Outlier AI
OtherTextRLHF
I trained and evaluated large language models using RLHF, safety alignment, and rubric-based evaluation. I crafted domain-specific prompts and reference responses for creative, business, and technical tasks. I conducted model response ranking, safety checks, and contributed structured feedback to improve AI performance.
• Developed and applied detailed rubrics for evaluation.
• Conducted side-by-side response comparisons and structured feedback loops.
• Participated in AI safety and red-teaming projects to flag harmful or biased outputs.
• Used JSON and prompt engineering for business and computer science-related model tasks.
I trained and evaluated large language models using RLHF, safety alignment, and rubric-based evaluation. I crafted domain-specific prompts and reference responses for creative, business, and technical tasks. I conducted model response ranking, safety checks, and contributed structured feedback to improve AI performance. • Developed and applied detailed rubrics for evaluation. • Conducted side-by-side response comparisons and structured feedback loops. • Participated in AI safety and red-teaming projects to flag harmful or biased outputs. • Used JSON and prompt engineering for business and computer science-related model tasks.
2025 - Present