Share Your Story
Share Your Story
Artificial Intelligence November 20, 2026

Apple Is Hiring Engineers to Build Its Own AI Agent That Can Use a Mac Like a Person

Apple is hiring engineers to build its own AI agent that can browse the web, work across apps and write code, with a possible arrival by the end of 2027.

Apple has posted job listings that show it is building its own AI agent, one that can use a computer on its own. The listings appeared on the company’s careers site this week and were first reported on 19 November.

The role at the centre of the story is Senior Engineering Manager for Agentic AI in Cupertino. The listing describes a proactive assistant that works across apps and devices. Apple has not announced a product, so the hiring post is the clearest public sign of its plans yet.

What the listing says

“Imagine an AI agent that can use a computer the way a capable person does. It browses the web, works across apps, runs commands, writes and debugs code, and calls services, and it chooses the right tool for each step.”

The team behind the work brings together engineers, applied AI researchers and systems specialists. Their agents are meant to plan and carry out tasks that run across several steps, moving between apps, browsers, command lines, code and programming interfaces. Smaller specialised agents would handle parts of a bigger job.

The listing also says the new hire will work with Apple’s Siri, user interface and Apple Foundation Models teams. That places the agent inside the same organisation that builds Siri and Apple’s on-device models.

Why Apple is moving now

The agent race has moved fast in the past year. Meta’s Muse, OpenAI’s Dots and Anthropic’s Claude can already take actions on a computer, and Google has just given Gemini the ability to use a Mac. Apple has watched rivals ship while its own assistant has stayed mostly reactive.

The timing is striking because Google’s move landed in the same week. Gemini can now operate a Mac on the user’s behalf, which puts pressure on Apple to answer with an agent of its own. The company is late to this race, and the listings show it knows that.

Privacy is Apple’s opening

Apple has one advantage its rivals cannot copy easily. Its agent would run inside its own operating systems, with deep access that outside agents have to ask for. The listing stresses that the work will be done privately, which fits the company’s long pitch on privacy.

Apple is also tightening the rules for everyone else. The macOS 27.3 beta adds new controls that let users decide which apps and AI agents can see their files and folders.

The next iOS release lays the groundwork for Siri to work with outside apps, so it could book a car or a restaurant table. Apple is building its own agent while making sure rival agents need clearer permission first.

What happens next

The listings suggest the work is still at an early stage. Apple has talked about introducing the agent as soon as the end of 2027, but it has confirmed no date and no name. Hiring a senior leader now suggests the foundations are being laid rather than finished.

The push comes from the top. New chief executive John Ternus has been pressing software chief Craig Federighi to move faster and take a bolder approach to AI. The agent is a central part of that effort, and investors noticed at once.

Shares in Meta fell nearly 2 percent in late trading on 19 November, and Anthropic shares dropped more than 3 percent after the report appeared. The market read Apple’s hiring as a real threat to the agent makers that are already in the field.

Apple is late, but it is not starting from nothing. It owns the Mac and the iPhone, and an agent built into those devices would reach hundreds of millions of people on day one. The open question is whether it can ship before the agent race moves on without it.