← Back to Directory

Meta interview notes, page 63 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·Software Engineer·Technical Phone Screen

May 2026
Meta coding round focused on a grid BFS variant with a directional chute cell mechanic. Not a hard algorithm problem, but the direction logic is fiddly enough that it's easy to trip over if you're not careful about your representation choices.
  • Extend a grid BFS to handle a special one-way 'chute' cell where you can only enter from one side and must exit from the opposite side. How do you model this and ensure invalid traversals are rejected?

“The algorithm part is fine, standard BFS, not the issue.”

View Post

Meta·Software Engineer·Technical Phone Screen

May 2026
Meta coding round, one question the whole time. It's a string parsing problem that looks deceptively manageable until you're actually writing the recursive logic under pressure.
  • Given a string encoding of a binary tree where each node is its value followed by optional parenthesized left and right subtrees, parse the string and construct the actual tree. Handle single nodes, multi-digit values, and missing right children.

“I got the basic structure pretty fast but the index tracking killed me.”

View Post

Meta·Data Scientist·Technical Phone Screen

May 2026
SQL-heavy Data Scientist screen at Meta, four questions all centered on a shop visibility dataset with window functions, cohort analysis, and a stats reasoning piece at the end. Nothing behavioral, just raw SQL the whole way through.
  • Using a shop visibility table with timestamps, find the top 10 profiles by number of visibility flips (transitions where visibility changes from the previous state), breaking ties by smaller profile_id first.
  • Calculate the percentage of shops currently visible as of a given date, where 'currently visible' means the shop's last recorded state at or before end of day is visibility=1. Return both the percentage and raw counts.
  • Break down currently visible shops by category: for each shop_category, return the count of visible shops and its share of all currently visible shops.
  • Verify the hypothesis that newer shops are visible for fewer days. For each creation month cohort, compute each profile's fraction of days with final visibility=1 in its first 30 days since creation (deduping consecutive identical states), then return cohort-level profile count, median fraction, and P75/P25. Also explain how you'd handle profiles created less than 30 days before the analysis date, and how you'd test the trend statistically.

“LAG over ts partitioned by profile_id is the move here.”

View Post

Meta·Software Engineer·Onsite - System Design / Architecture

May 2026
Meta system design round, one question about building a people recommendation feature. Pretty open-ended, which I wasn't fully prepared for.
  • How would you design a 'people you may know' recommendation system?

“I jumped straight into graph traversal and mutual friends because that felt obvious, but then the interviewer kept pushing on scale and I realized I hadn't thought about how you'd actually compute this for hundreds of millions of users without it being a disaster.”

View Post

Meta·Software Engineer·Technical Phone Screen

May 2026
Meta SWE coding round with a string manipulation problem. Pretty straightforward premise but the edge cases around what counts as a valid reversal tripped me up a bit.
  • Given a string, you can reverse a prefix or a suffix of it to produce new strings. Iterate over all such possible strings and return the lexicographically smallest one.

“My first instinct was to just brute force every prefix and suffix reversal and track the minimum, which works fine for small inputs.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

May 2026
Meta product sense round, news app feature design. One question, pretty open-ended, the kind where you can either nail the structure or just ramble for 30 minutes and call it a day.
  • How would you design a feature to surface trending topics in a News app?

“I went straight to defining what 'trending' even means, which I think saved me.”

View Post

Meta·Software Engineer·Onsite - System Design / Architecture

May 2026
Meta system design round, got a reservation system question (think Ticketmaster or hotel booking). Pretty classic but the contention angle made it genuinely tricky to get right under pressure.
  • Design a reservation system where users can search availability, place a temporary hold on a specific seat or room, and complete checkout without overselling inventory. Holds should expire automatically, and the system needs to handle high contention for popular events.

“The hold expiration part is where I spent too long.”

View Post

Meta·Machine Learning Engineer·Onsite - Behavioral / Leadership

May 2026
Behavioral round at Meta for an ML Engineer role. Just one question but they really sat with it, lots of follow-up probing on the specifics.
  • Tell me about a time you ran into serious obstacles while trying to move a project forward. What was the problem, how did you break it down, what did you do, and what happened?

“I had a decent story ready but the follow-ups caught me flat-footed.”

View Post

Meta·Machine Learning Engineer·Technical Phone Screen

May 2026
Went through a coding screen for an MLE role at Meta. One algorithm question, pretty focused on string manipulation. Nothing too wild but the two-pointer trick is easy to miss under pressure.
  • Given a string, return true if it can become a palindrome by removing at most one character.

“My first instinct was brute force, try removing each character and check.”

View Post

Meta·Software Engineer·Technical Phone Screen

May 2026
Meta data engineer interview that was basically a gauntlet of SQL and Python thrown at you in one sitting. Heavy on the engineering side, almost no soft skills stuff, which I wasn't fully expecting.
  • Given a clickstream events table and a users table, write SQL to compute DAU, WAU, and MAU.
  • Using the same schema, write SQL for D1 and W1 retention cohorts.
  • How would you sessionize clickstream events using SQL, and what metrics would you compute per session?
  • Design an incremental daily job that updates aggregation tables idempotently. How do you use MERGE or INSERT patterns alongside window functions to make this work?
  • How do you detect and handle duplicate events and late-arriving data in a pipeline?
  • Write a Python script (no heavy frameworks) that parses semi-structured JSON from the properties column, normalizes nested fields, and writes the output as partitioned Parquet files.
  • How would you write basic unit tests for the data-cleaning script you just described?

“Seemed straightforward at first and I kind of rushed it.”

View Post