Hiring nowTool-using agents and workflow orchestrationPart of The agent workforce

AI Agent Engineer

Also called Agentic AI Engineer · Agent Engineer · AI Agent Developer

An AI Agent Engineer builds software agents that take real actions through tools they are explicitly permitted to use. They turn model capability into automated work a company can trust, observe and switch off.

The Agent WorkforceAI Agent Engineers: hire for shipped agents, not for promptsRead the analysis: what agents already do in this role, and who to hire

What they do

The work is building agent services that plan a task, call tools, hold state and recover when something fails. Day to day, that means writing tool contracts with clear inputs, outputs and permissions, designing planning loops, and deciding what the agent may and may not do. It also means adding logging so every action can be traced, building sandboxes for testing, and writing task evaluations that show whether the agent finished the job. A good deal of time goes on failure: timeouts, bad tool output, half-finished tasks and handing over to a person.

What a strong hire delivers

Within the first months, a strong hire ships one bounded agent into real use. It has written permission boundaries, traceable actions and a task evaluation set the team can rerun. They give the company a clear view of authorised task completion, how often people must step in, and where failures cluster. They also leave reusable tool contracts behind for the next agent.

How you can tell they are good

Ask them to walk through an agent they shipped, not a demo. Strong candidates explain the permission model, how state was kept, what failed in production and how they measured it. Ask how they decided where a human steps in. Worry if they talk only about prompts and frameworks, or cannot say how they tested the agent.

Why it is hard to fill

Titles are fragmented, so the same work appears as agent engineer, AI developer or simply software engineer. Genuine production experience is rare because most agent work is new and much of it never left prototype. Candidates mix application engineering, integration and security thinking, and few have all three. Searching by title misses most of them.

Where to find them

Most come from application engineering, workflow automation or systems integration, then moved into LLM work. Look at engineers who have built with agent frameworks, designed tool or protocol integrations, or run workflow engines. They sit in AI startups, SaaS companies and enterprise automation teams. Search on shipped agents and framework contributions, not job titles.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for an AI Agent Engineer

  1. Describe an agent you shipped. What actions could it take, and who decided its permissions?

    A strong answer names specific tools, explicit permission limits and who signed them off. It covers what changed after launch. Worry if permissions were broad by default, or if the candidate only describes a prototype.

  2. How do you design a tool contract so a model uses it correctly?

    Look for narrow tools, typed inputs, clear errors the model can act on, and idempotent behaviour where possible. Worry about vague tool descriptions, one tool that does everything, or no thought for misuse.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building agents that carry out real tasks for our customers, inside clear limits. We need an engineer who can design, ship and measure them. The work covers tool design, planning, state and recovery, with close attention to what each agent is allowed to do.

What you will do

  • Build and ship tool-using agent services that complete defined tasks
  • Design tool contracts and permission boundaries for each agent
  • Implement planning loops and state management that survive failures
  • Add tracing so every agent action can be reviewed
  • Create task evaluations and track completion, intervention and failure rates
  • Design recovery paths and handovers to people
  • Work with product and security on safe rollout

What you will bring

  • Shipped at least one agent or tool-using LLM workflow into real use
  • Strong Python or TypeScript engineering
  • Experience with LLM APIs and agent frameworks or SDKs
  • Has designed tool or protocol integrations, such as MCP
  • Can explain how an agent was tested and where it failed
  • Sound judgement on permissions and security risks

Nice to have

  • Background in workflow automation or systems integration
  • Experience running agents in sandboxed environments
  • Experience with workflow engines

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Agent orchestrationAdvancedCoordinate tool-using agent execution
  • Tool and protocol integrationAdvancedConnect models to external data and capabilities
  • Security engineeringWorkingProtect systems, identities and sensitive assets
  • Model evaluationWorkingMeasure capability and failure modes credibly

Other roles in tool-using agents and workflow orchestration

Hiring nowAgent Tool Integration EngineerAn Agent Tool Integration Engineer connects AI agents to outside systems through tightly scoped interfaces. They make sure agents act only where…EmergingAgent Workflow ArchitectAn Agent Workflow Architect decides which tasks go to agents, which stay with people, and where approvals sit. It matters because agents that act…EmergingComputer-Use Agent EngineerA Computer-Use Agent Engineer builds agents that operate real application interfaces, clicking, typing and reading screens, within firm limits. They…EmergingAgent Memory and Personalisation EngineerBuilds the memory, user models and personalisation layers that let an agent remember people across sessions and act on it. It matters because an…EmergingAI Agent ManagerAn AI Agent Manager assigns work to AI agents, reviews what they produce, handles the exceptions and tunes the workflow. They make agents dependable…

Related roles in other families

EmergingMulti-Agent Systems ResearcherMulti-agent coordination and economicsA Multi-Agent Systems Researcher studies how several AI agents behave when they work together, and builds methods and benchmarks to make that…EmergingAgent Coordination EngineerMulti-agent coordination and economicsAn Agent Coordination Engineer builds the layer that lets several AI agents hand work to each other, share state and settle conflicts. It matters…EmergingMulti-Agent Evaluation ScientistMulti-agent coordination and economicsA Multi-Agent Evaluation Scientist measures how groups of AI agents behave together and designs tests that expose collusion, cascades and…EmergingAgent Market Mechanism ResearcherMulti-agent coordination and economicsThis person designs and tests the rules that govern how AI agents trade, bid and share resources in bounded simulated markets. It matters because…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Tool-using agents and workflow orchestrationAll new job families

Start a search

Hiring an AI Agent Engineer?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.