About the role
Xebia CEE is hiring a Senior AI Engineer to join a product team building a voice-first AI agent that supports natural spoken conversations. The agent reasons, calls tools, and takes actions in real time, and the product is still in active development, so architecture and direction remain open to influence. This is a hands-on role spanning the full voice-agent stack, from real-time communication and speech processing to orchestration, memory, and backend integrations.
You will design agent workflows with LangGraph, covering state management, tool calling, multi-step reasoning, memory, error recovery, and human handover. On the voice side, you will tune real-time pipelines using Pipecat, LiveKit Agents, or similar, working with streaming STT and TTS, voice activity detection, turn-taking, interruption handling, and latency reduction. The role also covers architecture decisions, evaluation.
Highlights
- Build a production voice-first AI agent using LangGraph and real-time voice pipelines
- Own architecture, evaluation, observability, and latency optimization across the stack
- Work remotely from the EU with in-person team meetings every 4 to 6 weeks
- Collaborate closely with product, design, and a small engineering team
This recap is dataskew's editorial summary, not the company's copy.