← Back to Directory

Meta interview notes, page 104 of 238

FAANG / Big Tech

Most of these Meta write-ups come from Software Engineers (1,194), then Data Scientists (489) and Product Managers (453). The most reported rounds are the technical phone screen (1,190), the product sense round (466), the system design round (255) and the behavioral / leadership round (205).

2377 interview notes · updated Jul 2026

Meta·Product Manager·Onsite - Product Sense / Strategy

May 2026
Meta PM interview with a product design question focused on a specific user segment. Pretty open-ended, which sounds nice until you're actually sitting there trying to figure out where to start.
  • Design a product or feature for new or expecting parents on Facebook.

“I spent too long in the user segmentation phase and kind of lost the thread.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

May 2026
Meta PM interview with a monetization question that sounds straightforward until you actually try to answer it with any rigor.
  • How would you increase per user spending by 25% on an e-commerce platform?

“I jumped straight into personalization and upsell tactics and felt pretty good about it, but looking back I skipped the part where you actually interrogate the number.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

May 2026
PM interview at Meta with a single product design question about audio. Pretty open-ended, which I wasn't fully prepared for.
  • How would you design an audio product for Facebook?

“I went straight to podcasts and voice messaging and kind of stayed there too long.”

View Post

Meta·Software Engineer·Technical Phone Screen

May 2026
Meta coding round, linked list reordering problem. Pretty standard stuff if you know the pattern, but I fumbled the implementation a bit under pressure.
  • Given the head of a singly linked list, reorder it so that nodes alternate between the front and back of the list (first, last, second, second-to-last, and so on), modifying only the pointers.

“I knew the general idea going in but blanked on how to cleanly split the list.”

View Post

Meta·Machine Learning Engineer·Technical Phone Screen

May 2026
Meta MLE technical screen focused entirely on graph traversal and topological sort. Pretty standard stuff for the role but they pushed hard on the tradeoffs discussion at the end, which I wasn't fully prepared for.
  • Given a directed graph of package dependencies as an adjacency list, implement a function that returns a valid installation order for a list of packages, or signals that no valid order exists if there's a cycle.
  • Walk through the tradeoffs between iterative and recursive DFS for this problem, including stack overflow risks and complexity.

“I jumped straight to DFS which was the right call, but fumbled the cycle detection part for a minute.”

View Post

Meta·Machine Learning Engineer·Onsite - System Design / Architecture

May 2026
Meta MLE system design round focused on ranking notifications from multiple sources into a unified feed. The cross-source value normalization angle was the real meat of it, and frequency capping turned into a pretty long rabbit hole.
  • Design a system to rank notifications from heterogeneous sources (friend activity, ads, system alerts) into a single unified inbox for each user.
  • How do you compare the value of an ad bid in dollars against a friend-event notification's predicted click-through rate? Walk through your normalization approach.
  • Walk through your approach to frequency capping across notification sources. How do you decide between a global per-user cap, per-source caps, and an adaptive cap?
  • How would you handle feedback loops where high-CTR notifications crowd out lower-CTR ones over time?
  • How would you handle cold start for a brand new user who has no behavioral history?

“This is the kind of question where you can spend 20 minutes on retrieval and never get to the interesting parts.”

View Post

Meta·Data Scientist·Technical Phone Screen

May 2026
Brutal onsite question at Meta for a DS role. One massive open-ended case covering metric design, observational analysis, experiment design, and long-term value quantification, all from two tables. The kind of question where you realize halfway through that you've been answering a slightly different question than they asked.
  • You have two event tables, one for feed views (with viewer, post, relationship type, duration, timestamp) and one for post reactions (like, comment, reshare, follow, hide, report). Your hypothesis is that content from friends drives more social engagement than content from unconnected sources. Design a full analysis end-to-end: define your metrics, validate the hypothesis observationally, design an experiment to test expanding unconnected content, and quantify the value of unconnected exposure even if near-term engagement is lower.

“This question ate me alive a little.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

May 2026
Meta PM interview, one question about defining success metrics for a business product. Pretty straightforward setup but the question has more surface area than it looks.
  • How would you define success metrics for Zoom as a business product?

“I started with user engagement stuff and the interviewer kind of waited, like they wanted more.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

May 2026
PM interview at Meta, one question about defining success metrics for Workplace. Pretty focused session, no fluff.
  • If you were the PM for Workplace, what metrics would you use to measure success?

“Went straight for engagement metrics and kind of forgot to anchor on what success even means for an enterprise collaboration tool versus a consumer product.”

View Post

Meta·Software Engineer·Technical Phone Screen

May 2026
Meta Research Engineer coding screen, one question on sparse vectors with a follow-up that actually made me think harder than the main problem.
  • Given two sparse vectors, design a class that stores them efficiently and computes their dot product. Follow-up: how does your approach change if only one of the two vectors is sparse?

“The main part is pretty approachable if you know to use a hashmap to store only the nonzero indices.”

View Post