AI Data Specialist
Freelance (Telus, Outlier, Stellar AI, Alignerr)
Oct 2024 - Current (1 year 11 months)
• Contributed to OpenAI's new agentic model development through comprehensive evaluation, implementing RLHF to optimise Large Language Model (LLM) efficiency and ensure compliance with quality standards • Evaluated AI model performance across safety, functionality, and reliability parameters, conducting adversarial testing and stress testing to identify model vulnerabilities and limitations • Created coding prompts and developed test cases including edge cases for model evaluation, assessing model prompts using comprehensive rubrics (sometimes 100+ pages long) • Marked, fixed, and evaluated others' work as a reviewer across multiple leading AI companies