Meta Engineering Manager (M1) Interview Experience
Meta · Engineering Manager
Not all interviewers may be able to adequately ask the right questions to go in depth, but they are absolutely looking for you to provide that depth by yourself. Your examples should not be surface level, but show you actually lived through the situation.
Interview process
I interviewed for an M1 Engineering Manager role, and the process included a screening round first, followed by the main loop. The screening consisted of one broad behavioral interview and one system design interview, in which I could choose between machine learning systems, software systems, and products; I chose machine learning.
In the main loop, I had another ML system design round, a project deep dive, a team management interview, another behavioral round, and an AI-assisted coding interview. The AI-assisted coding piece was the newest part of the process for me, but the bigger surprise was how behaviorally heavy the EM loop was.
My main takeaway is that they expect you to provide depth on your own, even if the interviewer isn't especially good at eliciting it.
Interview rounds · 7
- 1
Phone screen
BehavioralPeople ManagementCross-FunctionalI started with a one-hour screening behavioral round that covered a lot of leadership areas all at once, and it felt broad but still pretty detailed.
- Q1. How do you handle performance in your team, both high performers as well as managing out low performers?
Q2. Tell me about a time when your cross-functional partner gave you some critical feedback.
Follow-up questions- How do you deal with cross-functional stakeholders, including handling conflicts?
- Q3. How do you deal with growth and learning yourself, as well as critical feedback that you may have received from your manager or other stakeholders?
- 2
Technical round
System DesignMachine LearningProduct DesignThe screening technical round was an ML system design interview where I could choose between machine learning systems, software systems, and product, and I picked machine learning.
Q1. Design a recommendation system for a specific business use case.
Follow-up questions- Walk me through the requirements end to end.
- What data and features would you use here?
- What modeling approaches would you consider, and what are the trade-offs?
- How would you handle model training?
- What evaluation metrics would you use?
- What production concerns would come up?
- How would this scale?
- What edge cases arise when you move to more complex models like deep neural networks?
- 3
Technical round
System DesignMachine LearningProduct DesignIn the main loop, I had another ML system design round, and honestly it felt almost identical to the earlier one in terms of interviewer style and what they were evaluating.
Q1. Design a recommender system for recommending destinations on a travel website.
Follow-up questions- How does the different nature of the data change the architecture you would build?
- 4
Other round
BehavioralProject ManagementCross-FunctionalI expected the project deep dive to be me leading one project story for a long time, but it was actually much more of a situational back-and-forth.
Q1. Tell me about a time when you had to lead a large cross-functional project.
Follow-up questions- What was the most significant roadblock that you encountered during this project?
- How did you navigate conflicts?
- How did you handle ambiguity in the project?
- How did you deal with shifting timelines or delays?
- 5
Other round
BehavioralPeople ManagementThe team management round was centered on performance management and tested whether I could go deep on hard people situations, not just talk in generalities.
- Q1. What was the most challenging performance management that you had to handle?
- 6
Final / onsite round
BehavioralThere was also a separate behavioral round in the final loop, which reinforced how much weight they put on detailed leadership examples for EM candidates.
- 7
Technical round
CodingTechnicalArtificial IntelligenceThe AI-assisted coding round stood out because it was new and interesting, even compared with the rest of the loop.
Tips from the candidate
If I were prepping again, I would spend a lot of time on the behavioral side because for an EM it is extensive. There are three behavior-related rounds in the final loop and one in screening, so you need a lot of examples, and you need them in depth. I would not rely on the interviewer to guide me to the good parts of the story. I would come in ready to show that I actually lived through the situation, with specifics on conflict, performance, ambiguity, delays, and feedback, instead of giving surface-level answers.
What I took away is that Meta is evaluating EMs pretty holistically across multiple rounds, especially on leadership depth. The two ML system design rounds felt very similar in style, so it seemed they were treating both as data points rather than treating one as fundamentally different from the other. On the behavioral side, the bar felt high even when the interviewer wasn't great at asking the right follow-ups, because they still expected me to go deep without much prompting. The AI-assisted coding round also felt like a newer process element, so it seems like they are experimenting a bit while still keeping the core EM evaluation very behavior-heavy.