Machine Learning Architect - Conversational Speech

What you'll do

• Define technical direction and modeling strategy for speech recognition, synthesis, and dialog systems.
• Evaluate emerging research trends to develop actionable roadmaps for conversational AI technologies.
• Ensure architectural decisions are production-ready, considering constraints like latency and scalability.
• Provide deep technical guidance to multiple teams working on interconnected speech capabilities.
• Collaborate with cross-functional teams to integrate speech modeling investments into Apple's broader AI strategy.
• Champion the development of multimodal foundation models that integrate speech, audio, and text modalities.

What we're looking for

• 10+ years of machine learning experience in speech or multimodal systems with increasing technical leadership.
• Proven expertise as a modeling expert defining direction for multiple teams across interconnected speech capabilities.
• Deep hands-on proficiency in modern deep learning techniques including large language models and end-to-end speech systems.
• Significant experience building speech-to-speech conversational systems with full-duplex natural interaction understanding.
• Expert programming skills in Python and deep learning frameworks like PyTorch, JAX, or TensorFlow for production-quality systems.
• Track record of translating research into scalable, production-ready machine learning models.