AI and ML

AI Agents in AI Jobs

Short Answer, November 2026

In November 2026, 24.3% of the 2,179 technical job ads at 12 AI companies named AI Agents (529 ads), number 7 of the 40 skills ai2hired tracks. They are asked for most in Research roles (39.4%) and AI and ML Engineering (30.1%). In AI job ads in pharma, 31.4% name them.

  • 24.3%of technical ads at AI companies
  • #7of 40 skills tracked
  • 31.4%of AI job ads in pharma

The Skill Explained

What Are AI Agents?

An AI agent is a system where a language model works toward a goal in a loop: it decides the next step, calls tools such as search, code or business systems, looks at the results and continues until the task is done. Unlike a chatbot, it takes actions, not only answers.

Why AI Employers Ask for Them

Companies want AI that completes work, not only chats: triaging tickets, researching a question across sources, filling forms, fixing code. Building agents that are useful and safe is hard, so employers look for people who can design the loop, the tools, the limits and the checks. AI Agents are among the skills technical AI job ads ask for most.

How It Works

At the core is a loop. The model receives the goal, the history so far and a list of tools it may call, each described by a name, a purpose and the inputs it takes. The model replies either with a final answer or with a tool call. The application runs the tool, adds the result to the history and asks the model again.

Designs range from fixed workflows, where the steps are set in code and the model fills in each one, to open agents that plan their own steps. Several agents can also work together, each with a role. Frameworks such as LangGraph or agent software development kits help with state, memory and tool calling, and protocols such as the Model Context Protocol (MCP) standardise how tools are connected.

Reliability is the main challenge: an agent can loop, call the wrong tool or take an action it should not. Good agents limit what tools can do, log every step, ask a human before risky actions and are tested on realistic tasks.

What You Do With It at Work

  • Define tools (functions) an agent can call, with clear descriptions and safe permissions.
  • Design the workflow: fixed steps where possible, open planning only where needed.
  • Add memory and state so an agent can handle long tasks.
  • Trace and evaluate agent runs to find where they fail.
  • Put human approval in front of actions that cannot be undone.

Tools and Terms You Will Meet

Tool calling
The model asks the application to run a function and return the result.
Agent loop
Decide, act, observe, repeat, until the goal is reached or a limit stops it.
Workflow vs agent
Steps fixed in code versus steps the model chooses itself.
Human in the loop
A person reviews or approves certain steps.
Tracing
Recording every model call and tool call to debug and evaluate an agent.

Industry Example: Pharma

An agent prepares a first draft of a deviation investigation: it pulls the batch record, searches similar past deviations and the relevant SOP, and drafts the report. It cannot close the deviation; a quality specialist reviews and signs.

A Common Mix-Up

Not every chatbot is an agent. A chatbot answers; an agent takes steps and uses tools. Many valuable "agents" are actually simple, predictable workflows, and that is often the better design.

How Deep Do You Need to Go?

  1. Know itExplain the difference between a chatbot, a workflow and an agent.
  2. Use itBuild tool-calling workflows with logging, limits and human approval.
  3. Master itDesign reliable multi-step agent systems, evaluate them and secure their tools.

Start From Zero · 8 min

Your Road to AI Agents Starts Here

S0 · #1Learn AI to Get Hired: The Full Map

AI Agents has its own lesson later in the course, in Season 10. Every lesson builds on the ones before it, and the course starts with this free map of the whole road to an AI job.

Which Roles Ask for AI Agents

Share of each role family's job ads at AI companies that name AI Agents, November 2026.

  1. Researchtitles such as Research Scientist or Research Engineer, 188 ads39.4%
  2. AI and ML Engineeringtitles such as Machine Learning Engineer or Inference Engineer, 173 ads30.1%
  3. Software Engineeringtitles such as Software Engineer or Backend Engineer, 625 ads24.8%
  4. AI Security and Safetytitles such as Security Engineer or Trust and Safety Engineer, 157 ads24.8%
  5. Infrastructure and Platformtitles such as Infrastructure Engineer or Site Reliability Engineer, 165 ads24.2%
  6. Data Science and Analyticstitles such as Data Scientist or Data Analyst, 36 ads22.2%
  7. Solutions and Forward Deployedtitles such as Forward Deployed Engineer or Solutions Architect, 659 ads19.6%
  8. Product and Program Managementtitles such as Product Manager or Technical Program Manager, 153 ads19.6%
  9. Data Engineeringtitles such as Data Engineer or Data Platform Engineer, 23 ads8.7%

By Region

Share of technical ads naming AI Agents in each region.

  1. Remote28.1%
  2. North America25.7%
  3. Europe and the UK16.9%
  4. Asia and Pacific15.2%

In AI Job Ads in Pharma

31.4%of 1,768 AI job ads at 21 pharma and AI biotech companies, number 3 of 47 AI skills

  • Solutions and Consulting58.1%
  • Machine Learning and AI56.2%
  • Commercial, Medical and Market Access46.2%
The AI in Pharma index

Learn AI Agents for Free

Every ai2hired lesson exists because real job ads ask for the skill. Lessons build on each other, so nothing is used before it is taught.

How to Start, and How to Show It

  1. Build a model call with one tool (for example a calculator or a search) and handle its result.
  2. Turn it into a small workflow with 3 to 4 tools and log every step.
  3. Write 15 realistic tasks, run them, and fix the most common failure.

A CV Line Once You Have Done It

Built a tool-calling agent that drafts deviation reports from 3 internal systems, with step tracing and human approval; first-draft time fell from 3 hours to 25 minutes.

An example: use your own project, numbers and results.

Interview Questions on AI Agents

Two questions interviewers often ask, and what a good answer covers.

Q1

When would you build a fixed workflow instead of an open agent?

A good answer covers: When the steps are known: workflows are more predictable, cheaper and easier to test. Use open planning only where the path really varies.

Q2

How do you stop an agent from taking a harmful action?

A good answer covers: Limit tool permissions, validate tool inputs, require human approval for irreversible actions, set step limits and log every call.

Practice the full interview questions for each AI role

Questions About AI Agents in AI Jobs

What is AI Agents?

An AI agent is a system where a language model works toward a goal in a loop: it decides the next step, calls tools such as search, code or business systems, looks at the results and continues until the task is done. Unlike a chatbot, it takes actions, not only answers.

Why do AI employers ask for AI Agents?

Companies want AI that completes work, not only chats: triaging tickets, researching a question across sources, filling forms, fixing code. Building agents that are useful and safe is hard, so employers look for people who can design the loop, the tools, the limits and the checks. AI Agents are among the skills technical AI job ads ask for most.

Do AI jobs require AI Agents?

Often. 24.3% of technical job ads at AI companies named AI Agents in November 2026, and 39.4% of Research ads did.

Which AI roles ask for AI Agents the most?

Research (39.4%), AI and ML Engineering (30.1%), Software Engineering (24.8%). Shares are the part of each role family's job ads that name the skill.

Where can I learn AI Agents for free?

In the free ai2hired course: lesson 113, "Workflows vs Agents", in Season 10, which builds on the lessons before it. Season 0 is free on YouTube now.

Where do these numbers come from?

From the ai2hired AI Job Skills Index: every open role on the public job boards of 12 AI companies, read on 17 November 2026. A skill counts once per ad when the ad names it. The data is free to reuse under CC BY 4.0.

Cite or Embed This Data

Free to quote and reuse under CC BY 4.0, with credit and a link to this page. Data read on 17 November 2026; full edition: November 2026.

ai2hired, AI Job Skills Index, November 2026: AI Agents. https://ai2hired.org/ai-job-skills/agents/

Embed on Your Site

Paste this into a blog post or newsletter. The chart updates itself every month.

<iframe src="https://ai2hired.org/embed/ai-job-skills/agents/" title="AI Agents in AI job ads, ai2hired" width="100%" height="420" style="border:0;max-width:640px" loading="lazy"></iframe>
<p style="font-size:13px">Source: <a href="https://ai2hired.org/ai-job-skills/agents/">ai2hired AI Job Skills Index</a>, November 2026</p>

Learn AI Agents in Order, Free

Every ai2hired lesson is built from what real AI job ads ask for.