Agentic AI Engineer (human)

Your mission & challenges At NEURA Robotics, cognitive robots don't just execute commands. They perceive, reason, plan over long horizons, and act autonomously in the physical world. We're looking for an engineer who builds the systems that make that possible: the agent harnesses, planning loops, and orchestration layers that turn a language model into a reliable embodied agent.

• Agent architecture & harness design: Design and build the core agent harness for our humanoid platform: the control loop, tool/skill interfaces, state and memory management, retries, and guardrails that keep long-running autonomous behavior reliable.
• Long-horizon planning & orchestration: Build systems that decompose high-level human intent into multi-step plans, orchestrate skills and sub-agents, track execution state, recover from errors, and escalate to human oversight when risk is high.
• Conversational AI for embodied interaction: Design the real-time voice stack: STT → LLM → TTS pipeline with robust turn-taking, barge-in handling, and addressee detection so the robot knows when it's being spoken to and responds naturally under real-world latency constraints.
• Backend & platform engineering: Build the production backend behind the agent: well-designed APIs and interfaces, distributed services connecting sub-agents to the main conversational agent, and edge-to-cloud deployment across the NEURA ecosystem.
• Evaluation & industrialization: Stand up evaluation harnesses and observability that measure whether agents actually complete tasks correctly, and drive promising prototypes through to industrialized, deployed systems.
• Cross-functional collaboration: Work closely with hardware/software teams and external partners to take agentic capabilities from concept to reliable end-to-end product.

What We Can Look Forward To

• A strong Master's or PhD in Computer Science, Robotics, or a related field, or equivalent depth shown through shipped work.
• Demonstrated experience building production agent systems with measurable impact: agents that plan, use tools, maintain state, and run reliably over long horizons, rather than single-turn prompt work.
• Deep, hands-on experience with agent frameworks and orchestration such as LangGraph, AgentCore, MCP, or custom orchestrators, and a clear sense of when to use an agent versus a deterministic workflow.
• Professional backend engineering experience: production services, API and interface design, distributed and microservice architectures, clean code, testing, and DevOps discipline.
• Cloud and edge-to-cloud experience, including containerization (Docker, Kubernetes), CI/CD, and deploying and serving models and services in production.
• Strong grounding in LLMs, VLMs, and NLP.
• Excellent Python.
• Excellent English. German is a plus.