Senior AI Systems Engineer (Real-Time Voice + Agent Architecture)

We are building a high-performance AI operating system (LifeOS) with real-time execution.

This is NOT a chatbot project.

We are looking for a senior engineer capable of building a real-time AI pipeline with voice interaction and multi-agent routing.

CORE REQUIREMENTS:

1. Voice pipeline (REAL-TIME)
- microphone input
- streaming STT (speech-to-text)
- real-time processing through AI engine
- response generation
- streaming TTS (text-to-speech)

2. Agent routing (CRITICAL)
- ATHENA (Individual) ? must ALWAYS be used for Individual
- TARA (Ventures) ? must ALWAYS be used for Ventures
- NO cross-routing
- must be enforced at backend level (NOT UI)

3. Execution layer
- real actions (calendar, integrations)
- create / update / delete (real API execution)
- NO mock or fake execution

4. Performance
- real-time streaming
- low latency
- stable system

TECH STACK (flexible, but must justify):
- Python / Node.js
- FastAPI / WebSockets
- OpenAI / Claude
- STT: Whisper / Deepgram / Google
- TTS: ElevenLabs / Azure
- Redis / event-driven architecture

IMPORTANT:
- Senior level only
- You must have built real-time systems before
- No junior / no basic AI tools experience

TO APPLY — REQUIRED:
Answer these 3 questions:

1. How would you build a real-time voice pipeline (mic ? STT ? AI ? TTS)?
2. How do you enforce strict backend routing between multiple agents?
3. How do you reduce latency in a streaming AI system?

We are moving fast and expect production-level execution.

No partial delivery. Only complete systems.