- Min. Experience: 5+ years
- Type: Full time
- Work Mode: In-person
We're hiring an engineer to maintain and improve a production real-time AI voice platform.
What you'll do -
Own the backend services and the applied AI stack behind live voice and chat agents — speech-to-text, LLM orchestration with toolcalling, text-to-speech and retrieval. Improve latency, cost and reliability. Strengthen security and test coverage.
Requirements -
Strong async Python in production (FastAPI, asyncio)
PostgreSQL, Redis, AWS
Applied LLM engineering: reliable tool calling, structured output, prompt design, provider trade-offs
Production streaming STT and TTS integration
RAG you have shipped and debugged
Comfortable debugging systems you did not write.
Nice to have -
Real-time audio pipelines, telephony and WebSocket media streaming, LangChain/LangGraph, Hindi or another Indian language.
Clear writing and independent judgement matter here as much as code.