Research Scientist II – Deepfake Detection & Authentication (Video & Audio)

Pindrop is a cybersecurity firm focused on protecting interactions by confirming user identities through multiple channels including voice and video. The Research Scientist II role is embedded within the core research group, which develops advanced technologies to identify fraudulent or manipulated media and verify authentic users. The company supports major clients in sectors like finance and healthcare by analyzing vast amounts of interaction data to enhance security measures.

In this position, the scientist will independently spearhead projects aimed at creating and refining algorithms that detect manipulated audio and video content and verify biometric data. Responsibilities include building and testing machine learning models, ensuring research reproducibility, collaborating with engineering teams to implement scalable solutions, and investigating model weaknesses to improve reliability. The role also involves contributing to intellectual property and academic publications, requiring effective communication of complex technical concepts.

Applicants should hold an advanced degree or equivalent experience in a quantitative discipline, with a minimum of three years working on deep learning related to visual or audio data processing or biometric authentication. Strong skills in Python and frameworks like PyTorch or TensorFlow are essential. Experience in synthetic media detection and delivering research outcomes that translate into practical applications is required. This fully remote US-based role offers an opportunity to work on cutting-edge AI technologies that safeguard user interactions from sophisticated threats.