AI Model Language Specialist
ByteDance
Dec 2025 - Current (8 months)
Created and refined high-quality prompts and responses across diverse topics to improve LLM reasoning, clarity, and tone.
● Evaluated model behaviour across LLM and VLM systems, identifying risks related to bias, misinformation, and harmful content generation.
● Developed structured evaluation rubrics to assess hallucinations, bias, logical errors, and factual correctness.
● Acted as an adversarial user to identify model weaknesses, edge cases, and failure patterns.
● Performed rigorous fact-checking and validation of AI-generated content to ensure reliability and accuracy.
● Delivered detailed, actionable feedback to improve model performance and consistency.
● Maintained and updated annotation and evaluation guidelines for standardised qu