Reality Defender

The Applied Scientist II will be responsible for building, tuning, and deploying state-of-the-art audio deepfake detection models, focusing primarily on ensuring their robustness, reliability, and performance in diverse, real-world client environments. This involves investigating failure cases, building custom evaluation frameworks, and driving model iteration to handle various real-world conditions like compression artifacts and noise.

The engineer will design, build, and optimize ML/DL models for production-scale audio deepfake detection while interfacing with clients to understand production environments and performance criteria. Responsibilities also include investigating failures, building custom evaluation frameworks, and driving model tuning for performance under real-world conditions.