Principal AI Scientist

About Reality Defender

Reality Defender is a groundbreaking security and award winning (awarded most innovative startup at RSA 24) platform offering comprehensive deepfake detection. A Y Combinator graduate, Comcast NBCUniversal LIFT Labs alumni, and backed by DCVC, Reality Defender’s proactive deepfake and AI-generated content detection technology is developed by a leadership team with over 20 years of experience in applied research at the intersection of machine learning, data science, and cybersecurity.

With models defending against present and future fabrication techniques, Reality Defender is the best way to detect and deter fraudulent text, audio, and visual content, partnering with government agencies and enterprise clients to enhance security and detect fraud.

Role and Responsibilities

• Manage a computer vision team

• Deliver large portions of a project by defining timelines, model architectures, system design, and evaluation metrics for AI solution development and implementation.

• Identify complex but defined problems/gaps in existing technology and engage with cross-functional stakeholders and leaders to address them.

• Identify new and upcoming research areas for generative media detection by interacting with potential external and internal collaborators.

• Assist in research growth by sharing research trends and best practices internally and within the computer vision community by reviewing academic papers, and serving on program committees.

About You

• PhD degree in Computer Science or equivalent practical experience in Computer Vision.

• 4 years of experience with scientific agendas across multiple teams and projects for AI-first products.

• 2 years of experience in leading people and influencing scientific direction.

• 2 years of experience with cross-functional collaboration with product, engineering and business.

• Strong fundamental understanding of diffusion, GANs and visual generative AI techniques.

• Scientific publications at top-tier conferences, such as CVPR, ICCV and NeurIPS, especially on deepfake image/video detection and classification problems.