Key Insights
- The McKinsey Lilli interview is an AI-powered case interview that evaluates how candidates use AI, not just their final answer.
- The McKinsey Lilli interview is currently non-evaluative, but signals a shift toward AI fluency as a core consulting skill.
- Success in the McKinsey Lilli interview requires a structured workflow: Define the problem, guide the AI, challenge outputs, and synthesize a clear recommendation.
The McKinsey Lilli interview is a new AI-powered assessment the firm has begun piloting in select final rounds. While still evolving - and currently non-evaluative - this interview signals where consulting recruiting is heading.
For now, candidates are still assessed primarily on case interviews and the Personal Experience Interview (PEI). But the introduction of the McKinsey Lilli interview makes one thing clear: AI fluency is becoming a core expectation. Candidates who understand how to use AI effectively in problem-solving contexts will have a distinct advantage.
This guide breaks down what the McKinsey Lilli interview looks like today, what McKinsey is really testing, and how to prepare.
What Is the McKinsey Lilli Interview?
The McKinsey Lilli interview is best understood as a case interview with an AI copilot.
Instead of solving a business problem entirely on your own, you are given access to Lilli - McKinsey’s internal generative AI platform - and expected to use it as part of your analysis. You prompt the tool, evaluate its outputs, and incorporate those insights into your thinking while explaining your approach to the interviewer.
The key shift is subtle but important. McKinsey is no longer evaluating just your ability to solve a case. It is evaluating how you solve it when AI is part of the workflow.
Where It Fits in the Interview Process
Today, the McKinsey Lilli interview appears primarily in final rounds for select roles, particularly in the U.S. It does not replace the traditional case or PEI; instead, it is layered on top of them.
A typical final round may now include a standard case interview, a behavioral interview, and then this AI-enabled exercise. While the Lilli component is not yet heavily weighted, its presence signals what McKinsey will increasingly expect from candidates: The ability to operate in an AI-enabled environment from day one.
What Is Lilli?
Lilli is McKinsey’s proprietary generative AI platform, built on the firm’s internal knowledge base, including decades of research, case work, and proprietary insights.
In practice, it functions like a highly capable - but imperfect - analyst. It can generate ideas, structure problems, and provide directional analysis. But like any AI system, it can also be incomplete, overly generic, or occasionally wrong.
That tension is intentional. The McKinsey Lilli interview is designed to see how you handle it.
What the McKinsey Lilli Interview Looks Like
In the McKinsey Lilli interview, you are given a business scenario and asked to work toward a recommendation while using the AI tool.
You might begin by framing the problem and outlining an initial structure. From there, you use Lilli to explore problem drivers, generate hypotheses, or pressure-test assumptions. As the interaction unfolds, you refine your approach, challenge the outputs, and build toward a conclusion.
Throughout the exercise, the interviewer is less focused on what Lilli produces and more focused on how you engage with it. Do you prompt with intention? Do you question what you see? Do you synthesize effectively?
Candidates who treat the AI output as the answer tend to struggle. Candidates who treat it as input into their own thinking tend to stand out.
What McKinsey Is Really Testing
At its core, the McKinsey Lilli interview is still a consulting interview. The difference is that the same underlying skills are now being observed through a new interface.
McKinsey is looking for structured thinking - your ability to break down an ambiguous problem before jumping into analysis.
It is looking for judgment - whether you can recognize when an AI-generated answer is incomplete or misleading.
It is looking for iteration - how you refine your approach as new information emerges.
And it is looking for synthesis - your ability to step back and make a clear, defensible recommendation.
The presence of AI does not replace these skills. If anything, it makes them more visible.
How the McKinsey Lilli Interview Differs from a Traditional Case
In a traditional case interview, you are responsible for generating all of the analysis yourself. The process is relatively linear: you structure the problem, analyze it, and deliver a recommendation.
In the McKinsey Lilli interview, the workflow becomes more interactive. You are still responsible for structuring the problem, but you now guide an AI tool to generate parts of the analysis. That introduces a new layer: you must decide what to ask, how to interpret the response, and whether to trust it.
The result is a more iterative process. Instead of moving in a straight line, you cycle between prompting, evaluating, and refining. The challenge is not just solving the case - it is managing the problem-solving process itself.
McKinsey Lilli vs. Bain AI Interview: Key Differences Explained
McKinsey’s Lilli interview is the more developed of the two formats publicly today. It is already appearing in select U.S. final rounds, and reporting so far indicates it is currently a non-evaluative pilot layered on top of the standard case and PEI.
Bain, by contrast, has signaled that it plans to launch its own AI-enabled interview in Summer 2026, but has shared far less publicly about the exact format. What makes Bain notable is the firm's comment that the firm expects to see not only what candidates create with the tool, but also how they use AI in preparation - suggesting a broader view of AI fluency than McKinsey’s current pilot.
| Dimension | McKinsey Lilli Interview | Bain AI Interview |
|---|---|---|
| Status | Already appearing in select U.S. final rounds as a pilot. | Expected to launch in summer 2026. |
| Current weight in process | Currently described as non-evaluative and additive to the existing process. | Public weighting has not been disclosed. |
| Tool | Lilli, McKinsey’s proprietary generative AI platform. | Bain has not publicly named the tool yet. |
| Where it appears | Added alongside traditional final-round interviews such as case and PEI. | Likely to be incorporated into Bain’s interview process, but exact stage has not been publicly detailed. |
| What candidates do | Use AI during a live, case-style exercise to analyze a business problem and work toward a recommendation. | Use Bain’s AI tool during an interview exercise; exact structure has not yet been publicly described in detail. |
| What the firm is testing | Structured thinking, judgment, iteration, and synthesis while working with AI. | Likely similar core skills, but with additional emphasis on overall AI fluency. |
| Preparation scrutiny | Public reporting has focused on how candidates use AI inside the interview itself. | Bain has explicitly suggested it will have a record of how candidates use AI in their preparation. |
| What makes it distinctive | A more defined pilot format centered on collaborating with an internal AI copilot in real time. | A broader signal that Bain may evaluate both in-interview AI usage and pre-interview AI habits. |
| Candidate takeaway | Prepare to show that you can use AI without letting it lead the analysis. | Prepare not only to use AI well in the interview, but also to develop disciplined AI habits before it. |
The practical takeaway is that McKinsey currently looks more execution-focused, while Bain appears to be moving toward a more holistic assessment of AI behavior.
For candidates, that means the preparation strategy is slightly different. McKinsey candidates should focus on performing effectively in a live AI-enabled case.
Bain candidates should prepare for that too - but may also need to think more carefully about how they use AI in their broader interview prep, since Bain has signaled that this may become part of the assessment.
One nuance to keep in mind: Because Bain has not yet published a formal guide to its AI interview, some of the Bain side of this comparison is still inference based on public comments rather than a fully documented format.
How to Prepare for the McKinsey Lilli Interview
Preparation for the McKinsey Lilli interview is less about learning new technical skills and more about adapting your existing case interview approach.
First, the fundamentals still matter most. Strong candidates continue to excel at structuring problems, forming hypotheses, and communicating clearly. The AI component does not compensate for weak casing fundamentals.
Second, it is critical to practice using AI tools alongside case interviews. You do not need access to Lilli specifically; tools like ChatGPT or Claude are sufficient. The goal is to get comfortable structuring your thinking, translating that into prompts, and then critically evaluating the outputs you receive.
What differentiates strong candidates is not their ability to generate clever prompts, but their ability to stay in control of the analysis. They use AI to accelerate their thinking, not replace it.
Finally, develop a habit of synthesis. At every stage of the interaction, you should be able to articulate what you have learned and how it affects your recommendation. The interview still ends the same way: with a clear answer to the client’s problem.
Common Mistakes to Avoid
The most common mistake candidates make in the McKinsey Lilli interview is over-relying on the AI. When the tool becomes the driver of the analysis, it becomes immediately clear to the interviewer that the candidate is not in control.
A second mistake is failing to challenge outputs. AI-generated answers often sound polished, but that does not make them correct. Strong candidates pause, question assumptions, and test conclusions before moving forward.
Finally, many candidates fail to close the loop. They generate analysis, explore ideas, and interact with the tool - but never clearly state a recommendation. As in any case interview, synthesis is what ultimately matters.
Why McKinsey Introduced the Lilli Interview
The introduction of the McKinsey Lilli interview reflects how consulting work itself is changing.
Consultants are already using AI tools to research industries, generate insights, and test ideas. McKinsey’s goal is to ensure that new hires can operate effectively in that environment from day one.
The McKinsey Lilli interview is not just a new assessment - it is a simulation of how consultants increasingly work.
McKinsey Lilli Interview: Frequently Asked Questions
The McKinsey Lilli interview is an AI-powered case interview where candidates use McKinsey’s internal tool, Lilli, to analyze a business problem. The firm evaluates how candidates structure their approach, interact with AI, and synthesize insights - not just the final answer.
The McKinsey Lilli interview is currently non-evaluative in most pilot programs. It is used to test the format and assess candidate interaction with AI, while final hiring decisions still rely primarily on case interviews and the Personal Experience Interview (PEI).
The McKinsey Lilli interview differs from a traditional case interview by incorporating AI into the problem-solving process. Candidates guide AI-generated analysis, evaluate outputs, and iterate on insights, rather than performing all analysis independently.
To prepare for the McKinsey Lilli interview, practice case interviews while using AI tools like ChatGPT. Focus on structuring problems, writing clear prompts, challenging outputs, and synthesizing recommendations, as McKinsey evaluates judgment and process over technical AI skills.
McKinsey tests structured thinking, hypothesis-driven analysis, judgment, and synthesis in the Lilli interview. Candidates must also demonstrate effective AI usage, including prompting, iteration, and the ability to critically evaluate AI-generated outputs.
The Bottom Line
The McKinsey Lilli interview represents a shift in how candidates are evaluated, even if it is still early in its rollout.
The question is no longer just whether you can solve a case. It is whether you can solve it while effectively using AI without losing control of the analysis.
As the McKinsey Lilli interview continues to expand, candidates who understand how to use AI effectively will have a clear advantage. Those who treat AI as a shortcut will struggle, while those who treat it as a tool - and remain firmly in the driver’s seat - will be best positioned to succeed.