Voice AI Engineer

NeuroDrift is a US-based AI voice and enterprise software company. We build real-time voice AI agents that run live on real phone lines, at scale, for enterprise customers. Our platform (CallDash) handles production call traffic every day — so latency, telephony quirks, and audio edge cases are our daily reality, not a research problem. We're looking for a strong real-time voice engineer with broad expertise across the stack — someone who has shipped voice into production and debugged it when it broke under load.

What you'll do

• Build and operate real-time voice agents end to end — telephony, audio pipeline, LLM integration, tool calling
• Own latency: chase every millisecond from end-of-utterance to time-to-first-audio
• Integrate and harden SIP-trunk connections on real call platforms
• Deploy and scale stateful audio workers in production
• Take part in client meetings — present your approach and defend your technical decisions

What we need (3–4 years experience)

• 3+ years Python, including async (asyncio, FastAPI or similar)
• Real-time audio in production — WebRTC, SIP, RTP
• Production telephony integration — you've shipped at least one SIP-trunk integration on a real call platform and debugged it under load
• LLM voice integration in production — Gemini Live, OpenAI Realtime, ElevenLabs Conversational, Retell, or comparable
• Function / tool calling in voice contexts — you understand how tool-call scheduling affects perceived latency and barge-in behavior
• Kubernetes — deploying stateful audio workers with autoscaling
• A latency mindset — you instinctively reach for end-of-utterance, time-to-first-token, and time-to-first-audio metrics before the user complains

The setup

• Fully remote
• High-intensity role: 50–60 hours/week — this is startup pace, not a 9-to-5
• Working hours primarily IST, with availability for client meetings in US time zones
• You'll thrive here if you like owning hard problems end to end, you're comfortable in front of clients, and "it works on my machine" isn't in your vocabulary.