← Back to Directory

Meta interview notes, page 144 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

Jun 2026
Got a product strategy question at Meta about live streaming trends across different markets. Short on details from the original but the question itself was meaty enough to chew on for a while.
  • Live stream participation is declining in the US but growing in Spain. How do you approach this, and what would you do about it?

“This is the kind of question where you can spiral fast if you try to boil the ocean.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026
Meta SWE coding round, one question about buildings and sea views. Pretty classic stack problem once you figure out what they're actually asking.
  • Given an array representing building heights along a coastline, return the heights of all buildings that have an unobstructed view of the sea.

“Took me a second to realize the sea is on one side and you scan from that direction.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026
Meta coding screen, pretty standard algorithms territory. The whole thing centered on one problem but they pushed pretty hard on the follow-up optimizations and generalization, so it wasn't just 'write a loop and go home.'
  • Implement a power function that computes x raised to the integer n, where n can be negative. Walk through a naive solution first, then optimize it, and discuss edge cases.

“Started with the obvious loop and they let me finish before asking 'can we do better.' Binary exponentiation clicked for me pretty fast but I fumbled the negative n path for a second.”

View Post

Meta·Software Engineer·Onsite - Coding / Algorithms

Jun 2026
Meta SWE coding round with one algorithmic problem. Pretty straightforward simulation but the edge cases tripped me up more than I expected.
  • Given a non-negative integer array, repeatedly take the first non-zero element and subtract it from all leading elements that share the same value, continuing until all elements are zero or you hit an element strictly smaller than the one you're currently processing. Return the array at that point.

“I read it too fast and thought it was just a prefix sum thing.”

View Post

Meta·Software Engineer·Onsite - Behavioral / Leadership

Jun 2026
Behavioral round at Meta for a software engineering role. Two main areas: self-reflection on a recent failure, and a conflict resolution story with a follow-up about how the other person felt afterward. The follow-up on the conflict question is the one that'll catch you off guard if you haven't thought about it.
  • In a recent project, where could you have done better, and what specifically would you change if you did it again?
  • Tell me about a conflict you had with a teammate, partner, or stakeholder, and how you resolved it.
  • After that conflict was resolved, how do you think the other person felt about the outcome?

“The 'specifically' part is doing a lot of work here.”

View Post

Meta·Machine Learning Engineer·Onsite - Coding / Algorithms

Jun 2026
Meta MLE coding round, one problem that looked like a grid traversal warmup but kept growing. The bomb mechanic was the real test and I wasn't fully prepared for how state-heavy it gets.
  • You have a 2D grid maze with walls, open cells, a start, and an end. You're using BFS or DFS to find the shortest path. Now add a new cell type: bombs. When the traveler steps on a bomb, all wall cells within a blast radius of 2 (your choice of Chebyshev or Manhattan distance) are destroyed and become passable. Implement a helper that returns which cells get destroyed, and update your search so the planner correctly reasons about different possible detonation states.

“I went with Manhattan distance first and the interviewer was fine with that, just wanted me to document the choice.”

View Post

Meta·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Meta system design round, one question about designing the email infrastructure for Facebook. Pretty open-ended and I wasn't sure how deep to go on the backend vs the product side.
  • Design Facebook's email system.

“I spent the first few minutes just trying to scope it.”

View Post

Meta·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Meta SWE interview with a system design question on web crawlers. The setup was a distributed crawling problem and I'd seen a similar version floating around online, so I felt pretty prepared going in.
  • Design a distributed web crawler that can scale across many nodes crawling pages simultaneously.

“Had seen this one before so I wasn't going in blind.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026
Meta DS analytics round, one big open-ended question about comment engagement. The kind of question that sounds manageable until you realize how many directions it can go.
  • Given post, comment, and user tables, how would you analyze how comments are distributed across users to measure engagement? What metrics would you define, and how would you evaluate the impact of a new comment feature?

“I started with comments per DAU which felt right, but the follow-up pushed me toward distribution shape, not just averages.”

View Post

Meta·Data Scientist·Onsite - System Design / Architecture

Jun 2026
A deep system design interview at Meta for a Data Scientist role, focused entirely on building a hashtag recommendation system for News Feed end-to-end. The question was massive and covered everything from label construction to safety filters, so it felt less like one question and more like ten back-to-back. Brutal but fair.
  • How would you define the prediction target and unit of observation for a hashtag recommendation system in a social media News Feed? How do you construct positive and negative labels from logs without introducing selection bias?
  • What candidate generation sources would you use for hashtag recommendations, and how do you prevent popularity bias from dominating the results?
  • Describe at least ten concrete features you would use in a logistic regression ranker for hashtag recommendations, covering user affinity, post relevance, temporal signals, and locale.
  • Why start with a calibrated logistic regression instead of a deeper model? Walk through your regularization choices, how you'd handle class imbalance, and how you'd prevent data leakage.
  • How would you check and correct model calibration, and how would you set display thresholds per user cohort or surface? What about exploration rates for new hashtags?
  • What offline metrics would you use to evaluate the ranker, and how would you account for position bias in historical data when estimating top-k ranking quality?
  • How would you design the online experiment for this feature? Cover randomization unit, guardrail metrics, primary outcomes, novelty effect detection, and stopping criteria.
  • How do you handle cold start for new hashtags and new users, and how do you detect and respond to concept drift in hashtag popularity?
  • What safety and policy risks exist with hashtag recommendations, and how would you handle real-time filtering and fairness across languages and regions?
  • How would you translate the logistic regression coefficients into actionable product insights, for example around diminishing returns from showing too many hashtags or language mismatch penalties?

“This is where I spent the most time and probably the most words.”

View Post