AI Analyst Freelancer (RLHF, LLM Evaluation)
As an AI Analyst Freelancer at Deccan AI, I was responsible for evaluating large language model (LLM) code and logic via A/B testing based on technical rubrics. I contributed to the reinforcement learning from human feedback (RLHF) pipelines by generating high-quality ranking data and providing technical justifications. This ensured that model evaluations were rigorous and aligned with required criteria. • Evaluated code and logic produced by LLMs using task-specific rubrics • Authored technical reasoning and justifications for ranking outputs • Generated labeled data for use in RLHF pipelines • Employed remote workflows with technical documentation