AI Researcher & Evaluator (Freelance) performing RLHF/prompt engineering, multimodal evaluation/labeling, and factuality/safety assessments
Data Annotation Tech
VideoFreelance AI researcher and evaluator performing LLM capability tests, quality checks, and dataset improvement work across multimodal and text inputs. The role includes evaluating and labeling video and audio content, applying factuality/ground-truth verification, and executing red-teaming activities to identify biases, hallucinations, and unsafe outputs. It also involves prompt + response writing support and linguistic editing to ensure responses meet natural cadence, grammatical correctness, and cultural relevance. • Evaluate and label video/audio data for synchronization, sentiment, and intent recognition • Conduct ground-truth verification and fact-checking for medical and technical responses • Run red-teaming tasks for safety, ethics, and bias/hallucination mitigation • Edit and refine AI outputs and prompts to improve linguistic quality and constraint adherence