McKinsey Makes AI a Make-or-Break Skill in Job Interviews

McKinsey & Company has taken a decisive step into the AI era by making artificial intelligence a core part of its hiring process. In select final-round interviews, particularly for graduate roles in the United States, candidates are now required to use the firm’s internal AI tool, Lilli. The message is clear: the ability to work productively with AI is no longer a bonus skill—it is a baseline expectation.

According to CaseBasix, a US-based firm that prepares applicants for consulting interviews, McKinsey has rolled out what it calls an “AI interview.” In this new format, shortlisted candidates are given business problems similar to real client assignments and are expected to complete them using Lilli as part of their workflow. If an applicant struggles to engage with the AI tool or use it effectively, it can hurt their chances of being hired.

CaseBasix explained that the interview is designed to test how well candidates can work alongside AI rather than test technical expertise. “In the McKinsey AI interview, you are expected to prompt the AI, review its output, and apply judgment to produce a clear and structured response. The focus is on collaboration and reasoning rather than technical AI expertise,” the company said.

During these interviews, candidates are not judged on whether the AI provides the perfect answer. Instead, they are assessed on how they use Lilli to organise ideas, explore options, and sharpen their final recommendations. The exercise mirrors how consultants increasingly operate in real client environments, where AI tools are becoming everyday collaborators.

Importantly, McKinsey is not looking for AI engineers. Applicants are not expected to understand how Lilli is built or master advanced prompt engineering. As CaseBasix put it, candidates simply need to show they can use AI as a “productive thinking partner” and clearly explain how they arrived at their conclusions, much like consultants do when working with junior colleagues.

Early feedback suggests that the new AI interview focuses strongly on decision-making, judgment, and communication. As CaseBasix observed, “The McKinsey AI interview appears to assess how candidates think, judge and collaborate with an AI tool rather than their technical AI knowledge.”

This AI-based round does not replace McKinsey’s traditional assessments. It runs alongside two other interviews: one that evaluates problem-solving and structured thinking, and another that looks at leadership, personal impact, and values. Together, these rounds aim to give a more complete picture of how a candidate would perform in the firm’s increasingly AI-driven workplace.

The use of Lilli in hiring was first reported by the Financial Times, though McKinsey has not officially commented on the new interview format. Still, the firm’s leadership has been very open about how deeply AI is now woven into its operations.

McKinsey CEO Bob Sternfels recently revealed just how far this transformation has gone. In a Harvard Business Review IdeaCast episode, he said the firm’s workforce of around 60,000 people is supported by roughly 20,000 AI agents. Days later at the Consumer Electronics Show in Las Vegas, he updated that figure to nearly 25,000, a number McKinsey later confirmed to Business Insider.

That means more than one-third of McKinsey’s total workforce now consists of AI agents. Sternfels also highlighted the speed of this shift, noting that just a year and a half ago, the firm relied on only a few thousand AI agents. Looking ahead, McKinsey expects that within the next 18 months, every employee will be working alongside at least one AI agent as part of their daily routine—a future that new hires must now be ready to step into from day one.