AI Engineer

About Us: Tebi is creating the first Connected Restaurant Management System that makes hospitality flow. In one seamless platform, Tebi embeds sales, reservations, kitchen prep, payments, and all the data that powers it, creating smooth service and intuitive insights. Founded by Arnout Schuijff, co-founder of Adyen, and backed by €50 million in funding from Index Ventures & CapitalG, Tebi is built in service of hospitality first. Thousands of restaurants, cafés, bars, and coffee shops across the Netherlands, UK, Italy, Spain, Germany, Denmark and Finland already love running on Tebi.   About the role: We’re looking for a Senior AI Engineer to build and run the agents at the heart of Tebi OS. You’ll design, ship, evaluate and scale our Product Agents, along with the data pipelines, tools and infrastructure that make them reliable in production. The ideal candidate started in data or ML engineering and has since moved into AI engineering, building LLM-powered agents for real users. You enjoy turning messy financial and operational data into context an agent can reason over and act on with confidence and you are passionate about solving product challenges through AI. You’ll work closely with product and engineering to put intelligence into the hands of hospitality teams. That could be an agent that answers questions about a restaurant’s finances in real time, one that automates repetitive back-office tasks, or internal tools that help us move faster. What you’ll do: • Design, build and operate production agents: orchestration, tool and function calling, memory, multi-step reasoning, and integration with Tebi’s backend services • Build the data foundations agents rely on, including real-time and batch pipelines over financial and operational data, retrieval layers, and well-modeled datasets agents can query safely and accurately • Turn APIs and internal services into well-defined agent tools, and own the contracts between agents and the rest of the platform • Set up evaluation for agent behavior: offline eval sets, LLM-as-judge and human review, regression testing on prompt and model changes, and feedback loops that improve agents over time • Own observability for agents in production by tracing sessions, tool calls and model calls, and tracking latency, cost, error rates and quality • Work with product teams to find where agents can meaningfully improve the customer experience or internal workflows, and say clearly where they can’t • Make sure agents handle data securely and responsibly, with sensible guardrails, access controls, and protection against prompt injection and data leakage What you’ll bring: • 7+ years in software engineering, with a solid base in data or ML engineering • Hands-on experience building and shipping LLM agents to production, not just prototypes. You’ve dealt with flaky tool calls, runaway loops, hallucinated arguments and token budgets, and you have opinions on how to handle them. • Strong data modeling and data warehousing fundamentals, plus experience with ingestion, transformation and data quality • Strong SQL, and experience with pipeline tools (e.g. Airflow, dbt) and analytical or cloud data platforms (e.g. ClickHouse, BigQuery, Snowflake) • Experience evaluating and monitoring LLM systems, and an understanding of the full lifecycle from experiment to deployment to ongoing improvement • Backend engineering skills: you’re comfortable building the API endpoint that exposes an agent’s output to the frontend, or the service an agent calls as a tool • Clear communication, including the ability to explain trade-offs in latency, cost, accuracy and complexity to product partners • A product mindset. You care about how your work lands with users, and you’re keen to meet customers face to face t