Real-Time Voice AI Engineer — Sub-Second Conversational Pipelines

CloudCraft Labs LLC; Austin (TX); Hybrid; USD 130,000 - 180,000; Full time

Job summary

CloudCraft Labs LLC in Austin, TX is seeking a real-time audio engineer to build and optimize a streaming pipeline that connects speech-to-text, language models, and text-to-speech with sub-second response budgets. You will implement barge-in and graceful recovery to ensure human-like agent interaction.

Experience with Python and real-time media pipelines is essential, along with familiarity with STT/TTS/LLM APIs and related tradeoffs. Knowledge of WebRTC, SIP, or DSP is a Plus.

Qualifications

• Strong Python; experience with real-time audio, media pipelines, or streaming systems.
• Comfort with latency profiling and low-level performance work.
• Familiarity with STT / TTS / LLM APIs and their tradeoffs.

Responsibilities

• Build the real-time pipeline: streaming STT → LLM → TTS overlapped to a sub-second response budget
• Implement barge-in, turn-taking, and graceful recovery so agents feel human
• Own the WebRTC and telephony transports and the rehearsal sandbox
• Tune latency, prosody, and interruption handling against real call recordings

Skills

Python

Real-time audio

Streaming systems

Latency profiling

STT / TTS / LLM APIs

Tools

WebRTC

SIP

DSP

Job description

CloudCraft Labs LLC in Austin, TX is seeking a real-time audio engineer to build and optimize a streaming pipeline that connects speech-to-text, language models, and text-to-speech with sub-second response budgets. You will implement barge-in and graceful recovery to ensure human-like agent interaction.

Experience with Python and real-time media pipelines is essential, along with familiarity with STT/TTS/LLM APIs and related tradeoffs. Knowledge of WebRTC, SIP, or DSP is a Plus.