← Back to Directory

Meta interview notes, page 160 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·Data Scientist·Technical Phone Screen

Jun 2026
Meta DS interview focused entirely on ML system design, specifically building a restaurant recommendation system from scratch. The question was broad enough that you could go a dozen directions, which is either a gift or a trap depending on how you handle ambiguity. I think I did okay but definitely undersold the evaluation side.
  • Walk through the full design of a restaurant recommendation system end-to-end: what models would you use, how do you handle cold-start problems for new restaurants or users, and how do you evaluate recommendation quality both offline and in production?

“This one sprawls fast if you're not careful.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
Meta PM interview, product sense round focused on Maps. Single question, pretty open-ended, felt like they wanted to see how you think about niche user segments rather than just rattling off features.
  • You're a PM for Maps. What product would you build specifically for farmers?

“I blanked for a second because my brain kept defaulting to generic navigation stuff, which is obviously not the point.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026
A technical screen for a Data Scientist role at Meta that was basically a deep dive into a single fraud classifier scenario. Four sub-parts, all connected, and they expected you to carry context from one into the next. Not a casual conversation.
  • You inherit a binary fraud classifier. Given a holdout confusion matrix with TP=200, FP=800, FN=100, TN=99,900, compute precision, recall, F1, and false-positive rate.
  • False negatives cost $20 and false positives cost $0.20, with 1% fraud prevalence. How do you pick a threshold using predicted probabilities to maximize expected utility, and which metric, PR-AUC or ROC-AUC, better reflects improvements at low prevalence and why?
  • How would you check whether the classifier's predicted probabilities are well-calibrated, and how does calibration affect threshold selection?
  • Design an online evaluation plan that improves fraud catch rate without over-blocking legitimate users. Include guardrails, a shadow or staged rollout approach, success criteria, and rollback triggers.

“The numbers look clean but the class imbalance is jarring when you actually work through them.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026
Meta coding round, one tree problem the whole time. The question sounds straightforward but the O(1) space constraint is what actually makes it interesting, and I did not figure that out fast enough.
  • Given the root of a binary tree where each node has a 'next' pointer initialized to null, populate each node's next pointer to point to the node immediately to its right on the same level. Nodes with no right neighbor should have next set to null. Do this in O(n) time and O(1) extra space (no queue allowed). Return the root.

“My first instinct was BFS with a queue and I literally started coding it before they said no queue.”

View Post

Meta·Data Scientist·Onsite - Behavioral / Leadership

Jun 2026
Behavioral round at Meta for a Data Scientist role. The question was essentially one big multi-part prompt about conflict resolution under pressure, and it went deep fast.
  • Tell me about a time you resolved a major conflict within a team while under time pressure. Walk through the root causes you identified, the options you weighed, how you got buy-in from people who disagreed, how you kept the project on track, what the measurable outcome was, and what you'd do differently if there was a power imbalance or the team was remote.

“This thing had six parts and I did not clock that until I was already two minutes into my answer.”

View Post

Meta·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Meta ML engineer coding round, all grid traversal stuff. Four follow-ups stacked on top of each other, each one building on the last. Felt like a gauntlet more than a single question.
  • Given a 2D grid with walls, a start cell, and a goal cell, can you determine whether the goal is reachable from the start?
  • Now find the shortest path length between the start and goal in the same grid.
  • Reconstruct the actual path taken, not just the length.
  • Handle a variant of the grid problem, such as portals between cells, weighted movement costs, multiple goal cells, or finding all shortest paths.

“Standard BFS reachability check.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026
Meta SWE coding round, one question on interval merging. Pretty standard stuff but the edge cases are where they actually watch you think.
  • Given a list of intervals, merge all overlapping ones and return the result sorted by start time.

“The base case tripped me up more than I expected.”

View Post

Meta·Software Engineer·Onsite - Coding / Algorithms

Jun 2026
Meta SWE coding round focused on a substring-in-list problem, with the expectation that you'd walk through brute force first and then pitch progressively better solutions yourself. The AI assistant was available to implement, but the ideas and complexity analysis had to come from you. Felt more like a design conversation than a pure coding screen.
  • Given a list of strings, find a string in the list that contains at least one other string from the list as a substring. Then optimize your solution across two or three successive approaches, explaining time and space complexity for each.

“The AI could write the code but the interviewer wanted the optimization ideas to come from me, which I wasn't fully prepared for.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
PM interview at Meta focused on their AI assistant product. One question, pretty open-ended, but it has a lot of surface area if you're not careful about scoping.
  • Meta's AI assistant has been live for six months and is positioned as a core product experience across its family of apps. As a PM, how would you define goals for it over the next 12 months?

“This one sprawls fast.”

View Post

Meta·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Meta SWE interview that was supposed to be a coding round but turned into a full system design deep-dive. The question was dense and covered a lot of ground, from ingestion pipelines to ML classification to storage choices to APIs to experimentation hooks. Felt underprepared for the breadth of it.
  • Design a high-throughput content processing service for a large-scale media platform that can ingest tens of millions of items per day, handle bursty traffic, apply ML classification and rule-based filtering in a near-real-time pipeline with priority queues and backpressure, choose appropriate storage for media and metadata, expose APIs for upload, review, and appeals, support A/B testing and offline evaluation, and ensure scalability, fault tolerance, and cost efficiency. Provide a diagram and justify your trade-offs.

“I went in expecting a coding question so the pivot threw me for a minute.”

View Post