Voice AI Engineer

About the job Total Experience: 3+ Years Experience Building Voice AI: 1 year Location: Bengaluru Work Mode: In-office About Revspot Revspot is building AI-native revenue infrastructure for high-ticket B2C businesses. We are building AI agents that qualify leads, make voice calls, schedule meetings, engage customers, assist sales teams, and drive real revenue outcomes. We are looking for a Voice AI Engineer who has hands-on experience building and deploying production-grade Voice AI and LLM applications. This is a hands-on engineering role. You will build AI voice agents that can hold natural conversations, reason through customer interactions, call tools and APIs, and execute business workflows. What You Will Own Build and deploy Voice AI agents for lead qualification, customer engagement and sales use cases. Build real-time conversational pipelines across STT → LLM → TTS. Integrate voice agents with telephony systems, APIs, CRMs and internal tools. Build agentic workflows involving tool calling, function calling, memory, RAG and orchestration. Optimise voice systems for latency, interruption handling, turn-taking, accuracy and conversation quality. Build backend services and APIs required to power AI agents. Design evaluation and monitoring systems for voice-agent performance. Debug production conversations and continuously improve prompts, models and agent behaviour. Take features from problem statement → prototype → production. What We Are Looking For 3+ years of software engineering / AI engineering experience. Hands-on experience building Voice AI applications or conversational AI systems. Strong proficiency in Python. Experience building applications using LLMs and LLM APIs. Experience with at least some of these: OpenAI Realtime, Livekit, SIP Trunking, Langfuse, AWS Understanding of STT, TTS, VAD, turn detection, interruptions and real-time audio pipelines. Experience with APIs, WebSockets and asynchronous systems. Understanding of agentic workflows, tool calling and prompt engineering. Comfortable working with databases and cloud infrastructure. Ability to independently build and ship production features. Strong Plus Built a production Voice AI agent handling real customer calls. Experience reducing end-to-end voice latency and improving conversational quality. Experience with RAG, vector databases, embeddings, MCP or agent memory. Experience building evaluation frameworks for LLM/Voice AI systems. Experience with AWS and production backend infrastructure. Startup experience where you have built products from scratch. What Matters Most We are not looking for someone who has only experimented with LLM APIs. We want an engineer who has actually built AI systems and ideally has worked on Voice AI in production. You should be able to take a problem such as: "Build an AI agent that can call a lead, understand their intent, answer questions, qualify them…