Description:

Founding engineer on a conversational AI team, responsible for the real-time voice layer from speech input to spoken responses.
Focus on making natural voice interactions reliable in production with an emphasis on end-to-end latency.

Requirements:

Minimum 5 years of production software experience, with at least 2 years in voice, speech, or real-time audio systems.
Proven experience in building end-to-end real-time voice pipelines, including speech recognition and synthesis.
Proficient in Python or TypeScript, with hands-on experience in audio stacks like LiveKit or Twilio Media Streams.
Experience debugging audio at the frame level and building LLM evaluation harnesses.
Strong written English skills for asynchronous communication; experience with SIP or LLM orchestration is a plus.

Skills