Meta Interview Questions — Page 142

Meta·Software Engineer·Technical Phone Screen

Jun 2026

Meta SWE coding round with a pretty involved storage system design question. More implementation-heavy than I expected for a phone screen type setting, basically a mini system design baked into a coding problem.

Design and implement a two-level key-value storage system with expiry, prefix search, checkpointing, and rollback functionality.

“This one is deceptively large.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026

Meta SWE interview with a classic linked list problem. Not much to say, it was pretty standard.

Merge two sorted linked lists into a single sorted list.

“Pretty much a bread-and-butter problem.”

View Post

Meta·Software Engineer·Onsite - System Design / Architecture

Jun 2026

Meta system design interview, level 3 of 4 in some kind of integrated assessment. The problem was extending a work hours tracking system with promotion logic and salary computation. Fairly involved for a single question.

You have an existing work hours registration system. Extend it with two features: a promote() method that registers a promotion for a worker with a new position and hourly rate (where the new rate only kicks in after the current session ends), and a calc_salary() method that computes total earnings for a worker over a given time range by summing up the overlap of each completed session with that range, using the rate that was in effect at session start.

“The delayed promotion part tripped me up more than I expected.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026

Meta SWE coding round with a calculator problem that looks easy until you actually think about operator precedence and exponentiation. Pretty standard technical screen but the power operation tripped me up more than I expected.

Build a calculator that parses a string expression and evaluates it correctly, supporting addition, multiplication, and exponentiation, with proper operator precedence.

“I started with the easy case, addition and multiplication, and got through precedence fine using a stack.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026

Meta SWE coding round, got a tree traversal problem that sounds simple on the surface but has a few wrinkles if you're not careful about level-order logic.

Given a binary tree, return both its left side view and right side view. The left view is the first visible node at each level from the left, and the right view is the last visible node at each level from the right.

“BFS was the right call here and I knew it pretty fast.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026

Meta DS technical screen, all SQL, no behavioral at all. Four questions built on the same schema and they kept layering complexity. I came out feeling like I'd been run over by a window function.

Given a schema with users, friendships, posts, and feed impressions, classify each impression as coming from a friend or an unconnected author relative to the viewer. Then write SQL to compute, for each (user_id, impression_date), the fraction of impressions that are friend-sourced vs unconnected.
Define a weighted social engagement score per viewer-day and content source (friend vs unconnected), using like=1, comment=3, share=5. Compute the score from interactions on that date, where impressions with no interactions contribute 0. Return one row per (user_id, date, content_source) with impression count, interaction breakdown by type, and weighted score.
For a specific date, return the top 2 posts per user by weighted engagement score using the same weights. Use dense_rank() to handle ties so all tied posts at the cutoff are included, with ties broken by post_id ascending.
Compute month-over-month percent change in weighted engagement per content source between two consecutive months, aggregated across all users. Generate a month calendar CTE to handle missing months as zero before applying LAG. Explain your assumptions about multiple interactions on the same post and timezone boundaries.

“The join logic itself isn't hard but I kept second-guessing the direction.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026

Meta data engineer interview that went deep on metrics instrumentation and how you think about data quality end to end. The whole session felt like one long connected case study rather than separate questions, which I wasn't expecting.

For a new product feature, what would you define as the north-star metric and which guardrail metrics would you put in place alongside it?
Walk me through how you would instrument events for a new feature, build a tracking plan, and then derive KPIs from the raw event data.
How would you verify that your metrics are actually correct after instrumentation goes live, and how do you set thresholds for alerting?
If you notice an anomaly in a key metric, how do you investigate it?
How do metric definitions and their interpretation change when you're working with batch pipelines versus streaming pipelines?
How would you communicate data latency and metric caveats to non-technical stakeholders?

“I went straight to engagement as the north-star without really justifying why, and the follow-up pushed me on whether engagement was actually capturing user value or just activity.”

View Post

Meta·Data Scientist·Onsite - Behavioral / Leadership

Jun 2026

Behavioral round at Meta for a data scientist role, four questions all focused on how you work with people rather than anything technical. Pretty standard stuff but the cross-functional one had some teeth to it.

Tell me about a conflict you had with a teammate or partner. What was the disagreement and how did you resolve it?
Describe a situation where you used data or analysis results to change the mind of stakeholders who were initially resistant.
Walk me through a time you worked across multiple teams like product, engineering, or sales to get something done. How did you align everyone on goals and execution?
How do you handle it when urgent last-minute requests come in that conflict with work you already planned?

“I had a decent story ready but kept second-guessing whether my resolution sounded too passive.”

View Post

Meta·Software Engineer·Onsite - System Design / Architecture

Jun 2026

System design round at Meta for a Research Engineer role, focused entirely on integrating LLMs into a large-scale recommendation stack. Dense, technical, and honestly one of the more interesting interviews I've had in this space.

Why would you place an LLM at the final re-ranking stage rather than earlier in the retrieval or coarse-ranking stages of a recommendation pipeline?
How would you design a semantic ID scheme to tokenize items into a compact vocabulary of discrete codes for use with an LLM-based ranker?
How do you handle cold-start items that have no training history when using a semantic ID scheme?
What input and output format would you design for the LLM in this re-ranking stage, and would you use a sequence-to-sequence approach or a scoring head?
Given the strict latency requirements at the final ranking stage, what techniques would you use to make LLM inference fast enough?
How would you train this LLM ranker, and what role could reinforcement learning from human feedback play using engagement as the reward signal?
Describe the online serving architecture for this LLM-based ranker, including fallback paths and how you'd run an A/B test to evaluate it.
What are the main tradeoffs between using an LLM-based ranker versus classical approaches like gradient-boosted trees or mixture-of-experts models?

“This felt like the anchor question for everything else.”

View Post

Meta·Product Manager·Onsite - Behavioral / Leadership

Jun 2026

Meta PM interview, one behavioral question about project leadership. Pretty standard stuff but it still made me second-guess my story choice.

Tell me about a project you led that had a significant impact.

“Picked a project I thought was impressive but halfway through the story I realized I was spending too much time on the setup and not enough on what I actually did.”

View Post