Why We Replaced “AI Engineer” With “Agent Engineer”

We've been hiring AI Engineers in Berlin for months. Smart people. Outstanding résumés. Deep model knowledge. But something wasn't right.

They could fine-tune models. Craft prompts. Connect APIs. But when we asked about orchestration, state management or securing agent memory - most conversations went quiet.

We hadn't hired the wrong people. We had hired for the wrong role.

The job we had described as "AI Engineering" was really about designing, operating and securing the infrastructure for agents. And that discipline needed a name of its own.

We call it Agent Engineering.

It's still early. Agent Engineering as a recognised field is maybe a few months old. But Europe is only just waking up to it.

Arbio is one of the first companies in Europe to hire explicitly for this role.

What Agent Engineering Is

Agent Engineering is the practice of building, operating and managing agentic systems. It sits at the intersection of software engineering, system design and security engineering.

The focus is on:

• Runtime architecture: how agents are orchestrated, manage state and run concurrently
• Memory systems: how context, history and knowledge are stored
• Tooling integration: how agents are securely connected to APIs, databases or internal tools
• Safety and security: protecting data and enforcing access boundaries
• Evaluation and performance: tracking usefulness, latency and reliability

Agent Engineers don't just build agents. They build the systems that make agents useful, secure and scalable.

The Shift Is Already Happening

Every company rolling out AI products will eventually hire Agent Engineers - just as every software company eventually hired DevOps, then SREs, then Platform Engineers.

The pattern repeats because layers of abstraction always demand a discipline of their own. As systems become autonomous, the engineering challenge shifts from what they do to how they run.

We're early. Agent Engineering as a recognised field is maybe a few months old. But demand is growing exponentially. Europe is only just waking up to it.

Arbio is one of the first companies in Europe to hire explicitly for this role. Not because we chase trends. But because our infrastructure demands it.

What We're Building

We're building the runtime for property operations - agents that handle reservations, sync platforms, resolve conflicts and learn from the thousands of edge cases we've seen in our operations.

This isn't just another AI agent. It's a system. And systems need engineers who think in layers: agents, runtime, memory, security, observability.

If you want to build the thing that runs the thing - we should talk.

We're hiring Agent Engineers in Berlin.

For us, this isn't a title change. It's the recognition of a new engineering layer that will define how AI products are built and operated.