AI Trainer (contract) at Alignerr, remote
Performed AI model training using reinforcement learning from human feedback (RLHF). Evaluated, ranked, and assessed model responses for factual accuracy and overall quality. Specialized training work focused on air traffic control (ATC) use cases. • Conducted response evaluation and quality/ranking judgments • Supported RLHF training workflows • Focused on ATC domain application behavior • Ensured factual correctness and response quality criteria