AI Content Evaluator & RLHF Specialist (Large Language Models)
Scope: Worked as an AI Training Specialist focused on improving the accuracy, safety, and reasoning capabilities of a proprietary Large Language Model (LLM). Tasks Performed: RLHF (Reinforcement Learning from Human Feedback): Evaluated AI-generated responses based on strict criteria: Factuality, Relevance, and Instruction Following. Ranking & Comparison: Performed side-by-side (SbS) evaluations of model outputs, ranking them according to quality and safety guidelines. Prompt Engineering: Crafted complex prompts to stress-test the model’s ability to handle multi-step reasoning and creative writing tasks. Fact-Checking: Conducted external research to verify the truthfulness of model claims on high-stakes topics.