The role involves designing and building real-time streaming infrastructure to support bidirectional voice conversations with AI models. Responsibilities include developing low-latency serving systems, creating public and internal APIs, and managing audio transport layers to ensure smooth performance across various networks. The position requires building observability and quality-measurement systems to monitor latency and audio accuracy. Collaboration with research and product teams is essential to move model architectures from experimental stages into production environments. The work focuses on optimizing conversational responsiveness and ensuring that voice interactions remain reliable and natural for users.

Candidates must have at least 6 years of experience building distributed systems, real-time infrastructure, or platform services at scale. Proficiency is required in working across the technical stack, including transport protocols and serving infrastructure. A bachelor's degree or an equivalent combination of education, training, and experience in a relevant field is required. Strong communication skills and a collaborative approach to pair programming are essential for this role.