Audio AI Engineer

Zoom•San Jose, CA

3d•$137,700 - $275,400•Onsite

About The Position

As an Audio AI Engineer on Zoom's applied research team, you will advance the frontier of audio intelligence by developing and deploying machine-learning models that protect users from synthetic speech and audio deepfakes. Your work will have direct impact on platform trust and security, translating cutting-edge research into robust, production-ready systems. Each day brings a mix of independent experimentation, cross-functional collaboration with research, engineering, product, and security partners, and applied problem-solving at real-world scale. This position requires on-site presence five (5) days per week at our San Jose, California office. Remote or hybrid arrangements are not available for this role. Zoom's AI Research and Engineering team operates at the intersection of innovation and product, building foundational capabilities that power Zoom Workplace experiences for hundreds of millions of users worldwide. The audio AI group is focused on speech intelligence, security, and trust, ensuring that voice and video communications remain authentic and protected as generative technologies evolve. Your contributions here influence the integrity of Zoom's platform and help shape the broader industry conversation around responsible AI.

Requirements

PhD in Computer Science, Electrical Engineering, Machine Learning, Speech/Audio Processing, or a related field, or equivalent experience.
Research or industry experience in audio deepfake detection, synthetic speech detection, anti-spoofing, or speaker recognition; experience with generative audio, voice cloning, or adversarial machine learning is a plus.
Strong knowledge of speech and audio signal processing, with familiarity with large-scale or multilingual audio datasets a plus.
Deep understanding of machine learning, deep learning, model evaluation, and experimental design.
Proficiency in Python and deep learning frameworks such as PyTorch or TensorFlow; experience deploying and optimizing ML models in production is a plus.
Ability to conduct independent research and translate findings into practical, scalable solutions.
Publications in recognized speech, audio, AI, or machine learning venues are a plus; strong problem-solving, collaboration, and communication skills required.

Nice To Haves

experience with generative audio, voice cloning, or adversarial machine learning is a plus
familiarity with large-scale or multilingual audio datasets a plus
experience deploying and optimizing ML models in production is a plus
Publications in recognized speech, audio, AI, or machine learning venues are a plus

Responsibilities

Research, design, and develop machine-learning models for audio deepfake and synthetic speech detection.
Design training datasets, evaluation frameworks, and experimental pipelines for audio AI models.
Improve model accuracy and robustness across diverse speakers, languages, devices, codecs, and acoustic environments.
Analyze model failures and engineer solutions for reliable, real-world deployment.
Translate research prototypes into scalable, production-ready systems.
Collaborate cross-functionally with research, engineering, product, and security teams to align on priorities and ship impactful work.
Track and apply emerging developments in speech AI, generative audio, voice cloning, and deepfake detection.