← Back to Directory

Meta interview notes, page 174 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·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Meta ML engineer interview with a coding question on sparse vectors. Pretty short and focused, nothing wild, but the problem has a few gotchas depending on how you think about it.
  • Given two sparse vectors, compute their dot product efficiently.

“My first instinct was to just loop through both arrays, which works but completely misses the point of the word 'sparse'.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026
SQL round at Meta for a Data Scientist role. One question, pretty focused on ad analytics, joins, and window functions. Nothing too wild but the conversion rate edge case tripped me up a bit.
  • Given tables for ads, impressions, and conversions, write SQL to return total impressions, total conversions, conversion rate, and total revenue per ad for the past 30 days. Then extend it to rank ads by conversion rate and pull the top 10.

“The join chain was fine, impression to conversion is a left join and you group by ad_id, standard stuff.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026
A Meta DS interview that went deep on cross-app experimentation design, specifically around Instagram vs. Facebook ecosystem tradeoffs. The whole session felt like one long case study that kept spawning sub-questions every time I answered something.
  • How would you define a single overall evaluation criterion that captures healthy cross-app ecosystem value between Instagram and Facebook, without just rewarding one app cannibalizing the other's time?
  • You're measuring the impact of a new Instagram feature that might pull time away from Facebook. Walk through how you'd design the causal measurement plan, including your choice of randomization unit and how you'd handle cross-app interference.
  • Do a rough sample size and test duration calculation for this experiment given a baseline of 20 minutes per day on Facebook, an intraclass correlation of 0.02 at the geo level, a 1% minimum detectable effect on the ecosystem metric, 80% power, and a 5% significance threshold.
  • What guardrail metrics would you track alongside the primary OEC, and what would trigger you to stop the experiment early?
  • If the experiment results show Instagram's ecosystem metric improved but Facebook time dropped by 3%, how do you interpret that and what do you do next?

“This is where I stumbled first.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026
Meta DS interview with a product experimentation question centered on a sticker-reply feature for Facebook Groups. The whole thing was basically one big open-ended case that you had to structure yourself from objectives all the way through experiment readout.
  • Should Facebook Groups launch a sticker-reply feature? Walk through your hypotheses, success metrics, guardrail metrics, and how you'd design the experiment.
  • How would you analyze the experiment results and make a launch decision?

“This felt manageable at first and then immediately got slippery.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
PM interview at Meta, one product strategy question about Facebook Marketplace. Short and to the point, no idea how it went.
  • If you were the PM for Facebook Marketplace, what goals would you set?

“I went straight to growth metrics and kind of ignored the trust and safety angle, which I think was a mistake in hindsight.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026
Meta DS interview focused entirely on a ranking change question for a multi-account product feature. Two-part case: offline analysis first, then full A/B test design. Pretty deep for a single question but they really wanted to see how you think through the whole lifecycle from logs to experiment.
  • A product sorts a user's multiple accounts by most recent visit. There's a proposal to sort by number of notifications instead. Using historical logs and observational data, how would you evaluate whether this change is likely to be beneficial before running an experiment?
  • Design an A/B test to measure the causal impact of switching the account list sort order from most-recent-visit to most-notifications. Walk through unit of randomization, eligibility, metrics, sample size, and any pitfalls you'd watch for.

“This part tripped me up more than I expected.”

View Post

Meta·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
Meta PM interview, got hit with a classic metrics decline question. Pretty standard case format but the specifics of Facebook's social graph make it trickier than it looks.
  • You're a PM for Facebook. Friend requests are down 10%. What would you do?

“I jumped straight into diagnosis mode, which felt right, but I think I spent too long on the metric itself before clarifying what 'down' even meant.”

View Post

Meta·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Meta MLE coding round with a greedy array problem. Pretty clean problem once you see the trick, but I almost went with a brute force sort before the insight clicked.
  • Given two reward arrays of length n and an integer k, assign exactly k indices to mouse 1 (collecting reward1[i] for each) and the remaining n-k indices to mouse 2 (collecting reward2[i]). Return the maximum total reward.

“The greedy insight took me a minute.”

View Post

Meta·Data Scientist·Technical Phone Screen

Jun 2026
Brutal Meta DS interview focused entirely on search feature evaluation. Four interconnected parts, all building on each other, and I felt like I was barely keeping up by the third one.
  • You have two binary labels per query for a new search feature: relevancy and accuracy. Design a composite success metric using these labels. What's your exact scoring rule, how do you justify it under different error cost assumptions, and how would you calibrate the weights using actual business outcomes?
  • At what level do you aggregate this metric: query, session, user, or day? How do you handle users with many queries, missing labels, and the fact that outcomes within a session are probably correlated?
  • Design the online experiment for this search feature. Cover your randomization unit, which metrics are primary vs. guardrail, what inputs you'd use for power and sample size, how you'd pre-register the analysis, and how you'd prevent p-hacking when you're looking at many different slices.
  • Build out an offline evaluation pipeline: how do you collect labels, measure inter-rater agreement, maintain golden sets, and then monitor for label drift and Simpson's paradox when you segment results by query intent or locale?

“This is where I spent the most mental energy.”

View Post

Meta·Software Engineer·Technical Phone Screen

Jun 2026
Meta data engineer technical screen, one question, all SQL. The problem looked manageable at first glance but the ISO week logic and consecutive-week pairing made it a lot messier than a typical aggregation query.
  • Given a purchases table with customer_id, purchase_date, and rented_copies, write a single SQL query that: filters to calendar year 2024, uses ISO weeks (Monday to Sunday), finds customers who purchased in at least two consecutive full weeks, and for each such consecutive-week pair returns the customer with the highest total rented_copies in the first week of the pair, including ties.

“The part that tripped me up was defining 'consecutive' cleanly.”

View Post