Voice AI Backend Engineer

Description :

1. Speech-to-Text (STT)

2. LLM orchestration

3. Text-to-Speech (TTS)

4. Streaming audio response

- Integrate with telephony and IVR providers for inbound and outbound voice automation

- Implement WebSocket-based real-time audio streaming pipelines for low-latency voice interaction

- Work with WebRTC or media streaming architectures for real-time communication

- Integrate with voice model providers for :

1. Speech recognition

2. Streaming transcription

3. Neural text-to-speech

4. Voice-to-voice models

- Build scalable backend services to handle high concurrent voice calls

Optimize systems for :

1. Low latency

2. Streaming performance

3. Audio buffering

4. Real-time responsiveness

- Collaborate with product teams to build production-ready voice AI applications

- Design event-driven architectures for voice interaction pipelines

- Work closely with frontend, AI, and telephony teams to deliver end-to-end voice products.

Must-Have Criteria :

Core Requirements :

- Hands-on experience with voice bots / conversational AI systems

- Experience building GenAI solutions in production (LLMs, RAG, orchestration)

- Solid understanding of real-time audio streaming (WebSockets, streaming pipelines)

- Knowledge of voice AI architecture (STT ,LLM , TTS, turn management)

- Experience with telephony / IVR systems and call flows

Strong fundamentals in :

1. Distributed systems & microservices

2. Event-driven and API-driven architectures

3. Async programming & concurrency (Python)

4. Experience integrating voice/LLM model providers

5. Ability to debug latency and real-time streaming issues

Good to Have :

- Familiarity with WebRTC / LiveKit / real-time streaming frameworks

- Exposure to GPU/accelerator-based inference

- Experience in call center / voice agent systems

- Knowledge of VAD, interruption handling, multilingual voice systems

- Experience handling high-concurrency voice applications

Note : Only immediate joiners preferred or max 15 days