Research Scientist II – Deepfake Detection & Authentication (Video & Audio)
Core
Develop and deploy machine learning models for detecting AI-generated voice and video (deepfakes) and authenticating identities in real-time.
Role type
Senior IC research scientist (multimodal deepfake detection)
Builds
Production-ready deep learning systems for synthetic media detection and identity authentication
Domain
Cybersecurity, fraud prevention, biometrics, audio-visual processing
Deliverable
production ML models
Required skills
Deep learning, computer vision, speech/audio processing, Python, PyTorch/TensorFlow/Keras, reproducible research pipelines, model evaluation, bias/drift awareness
Preferred skills
C/C++ or model optimization tools (ONNX, TensorRT), real-time streaming pipelines, large-scale synthetic data generation, NLP, sequence modeling
Technologies
PyTorch, TensorFlow, Keras, ONNX, TensorRT, TF-Lite, Go
Responsibilities
Design and implement deep learning algorithms for audio-visual deepfake detection; develop internal research pipelines for training and benchmarking ML models; partner with engineering to transition prototypes to scalable production systems; analyze model behavior and attack vectors to improve robustness; contribute to patents and peer-reviewed papers
Seniority
Senior, hands-on IC
