AI Engineer - Conversational AI / Voice AI

About the Role
We are looking for an AI Engineer with 3–5 years of experience to build the AI layer of our real-time AI Call Assistant.
You will work on Voice AI, LLMs, Speech-to-Text, Text-to-Speech, multilingual conversations and real-time AI systems.
Responsibilities
Build real-time conversational AI for phone calls.
Integrate STT, LLM and TTS APIs, with AI.
Implement conversation flow, intent detection and caller memory.
Handle barge-in, silence, unclear speech and low-latency responses.
Support Telugu, Hindi and English.
Build call transcription and AI-generated summaries.
Optimize latency, reliability and AI response quality.
Implement AI guardrails and fallback handling.
Work with backend and Android developers to take AI features into production.
Requirements
3–5 years of experience in AI/ML, NLP, Generative AI, Conversational AI or Voice AI.
Strong Python skills.
Experience with LLM APIs.
Experience with STT/TTS or voice AI systems.
Understanding of real-time/streaming AI.
Experience with REST APIs and production AI systems.
Good problem-solving and experimentation skills.
Good to Have
Indian-language AI experience.
Twilio, Exotel or other telephony APIs.
WebSockets/WebRTC.
Telugu/Hindi speech AI experience.
Docker and cloud deployment.
Key Challenges You Will Work On
Making AI conversations feel natural on an actual phone call.
Reducing response latency between caller speech and AI response.
Handling caller interruptions and real-time turn-taking.
Improving speech recognition for Indian languages and accents.
Maintaining conversation context within a short call.
Building reliable AI fallbacks when APIs or networks fail.
Preventing hallucinations and fabricated commitments.
Creating measurable AI quality benchmarks for production.

Work arrangement
No