Forward Deployed Engineers (Voice AI) (Upto 30 LPA)

Role Overview

We are looking for an SDE 1 (AI) – Voice & Conversational AI to build and improve real-time AI agents that can listen, reason, interact with enterprise systems, execute tasks, and safely hand off conversations when required.

This is an engineering-focused role involving AI + backend systems + voice infrastructure, rather than prompt engineering alone.

Key Responsibilities

Build real-time voice agents with turn-taking, interruption handling, streaming, tool execution, and multilingual capabilities.

Integrate LLMs, speech-to-text, text-to-speech, telephony, APIs, databases, queues, and internal services.

Develop backend orchestration and state-management systems for multi-step AI workflows.

Build evaluation and observability frameworks covering task completion, latency, hallucination, reliability, and cost.

Optimise prompts, model selection, caching, retrieval, concurrency, and fallbacks.

Develop production-grade APIs and services with strong testing, error handling, retries, and monitoring.

Work closely with product and engineering teams to convert business workflows into reliable AI systems.

Required Skills

1–3 years of software engineering experience.

Strong programming skills in Python and/or TypeScript.

Hands-on exposure to LLMs, conversational AI, RAG, function/tool calling, or AI agents.

Good understanding of backend fundamentals, REST/WebSocket APIs, databases, queues, caching, Docker, and cloud deployment.

Strong debugging and problem-solving skills.

Ability to work in a fast-paced environment with high ownership.

Good to Have

Experience with voice/telephony systems such as LiveKit, Pipecat, Vapi, Retell, Twilio, Exotel, WebRTC, or similar.

Exposure to LangGraph, FastAPI, Node.js, Postgres, Redis, Kafka, vector databases, or Kubernetes.

Experience with multilingual or Indian-language conversational AI.

Projects, GitHub work, hackathons, or other demonstrable proof-of-work in applied AI.