Senior Voice AI Engineer

Job Summary

Senior Voice AI Engineer - Gurugram, India - Full-time.

We are looking for a Senior Voice AI Engineer to build and scale next-generation conversational Voice AI systems. You will work on real-time, low-latency voice pipelines involving Speech-to-Text (STT), Large Language Models (LLMs), Text-to-Speech (TTS), WebRTC, telephony, and streaming infrastructure.

You should have hands-on experience building production-grade Voice AI agents using frameworks like Pipecat, LiveKit, or similar real-time communication platforms. This role requires deep understanding of streaming architectures, WebSockets, voice quality optimization, interruption handling, and scalable backend systems.

Responsibilities

• Design and build production-grade Voice AI applications.
• Develop real-time streaming pipelines for voice conversations.
• Integrate STT, LLM, and TTS providers into low-latency conversational systems.
• Build robust WebSocket/WebRTC infrastructure for bi-directional audio streaming.
• Implement interruption handling (barge-in), turn detection, and Voice Activity Detection (VAD).
• Optimize latency across the complete voice pipeline.
• Integrate telephony providers such as Twilio, SIP, or LiveKit Telephony.
• Develop scalable backend services capable of supporting thousands of concurrent voice sessions.
• Design session management, state management, and conversation memory.
• Build monitoring, observability, and analytics for voice conversations.
• Deploy and operate Voice AI infrastructure on Kubernetes and cloud platforms.

Required Skills

Voice AI

• Strong understanding of conversational Voice AI systems
• Experience building real-time AI voice assistants
• Knowledge of latency optimization techniques
• Understanding of conversational memory and dialogue management

Speech Technologies

• Experience with OpenAI Realtime API, ElevenLabs, Deepgram, AssemblyAI, Google Speech, Azure Speech, Amazon Transcribe, Cartesia, or PlayHT
• Streaming STT and TTS
• Voice cloning and adaptive speech
• Speaker diarization
• Custom vocabulary and pronunciation dictionaries

Real-Time Communication

• Pipecat
• LiveKit
• WebRTC
• RTP
• SIP
• WebSockets
• Server-Sent Events (SSE)

Backend Development

• Python (preferred), FastAPI, AsyncIO
• WebSocket servers, gRPC, REST APIs
• Event-driven architecture
• Redis, Kafka, RabbitMQ
• PostgreSQL, MongoDB, ClickHouse (good to have)

AI LLM

• Experience with OpenAI, Anthropic, Google Gemini, or open-source LLMs
• Prompt engineering, tool calling, function calling
• RAG, agentic workflows, multi-agent orchestration, conversation memory

Cloud Infrastructure

• AWS, GCP, or Azure
• Docker
• Kubernetes
• NGINX
• Load Balancers
• Autoscaling
• CI/CD

Preferred Experience

• Built production Voice AI products
• Experience with Pipecat in production
• Experience with LiveKit Cloud or self-hosted LiveKit
• Experience with Twilio Voice or SIP infrastructure
• AI phone agents and call center automation
• Noise suppression (Krisp, DeepFilterNet, RNNoise)
• Experience supporting 1,000+ concurrent voice sessions
• Multilingual Voice AI

Qualifications

• Bachelors or Masters degree in Computer Science or related field.
• 6-8 years of software engineering experience.
• 3+ years building real-time streaming systems.
• 2+ years working on Voice AI or conversational AI products.
• Strong system design and distributed systems experience.

Disclaimer: This job description has been sourced from a public domain and may have been modified by Naukri.com to improve clarity for our users. We encourage job seekers to verify all details directly with the employer via their official channels before applying.