Meta Data Scientist Interview Experience — Offer on Christmas Eve After a Screen and a VO

Meta·Data Scientist·Oct 2025
OnsiteTechnical ScreenOfferhard

Got the offer on Christmas Eve — sharing it back with the forum.

Timeline:

  • Oct 29 — a university recruiter reached out
  • Nov 19 — screening interview with the interviewer
  • Dec 19 — VO (virtual onsite) with the interviewer
  • Dec 24 — offer

I was genuinely swamped at the end of the year, and on top of that I knew nothing about product case questions or SQL, so I ended up dragging this whole process out. Luckily I crammed hard right before the interviews and felt like it went okay overall. I had some other interviews lined up too, but honestly I'm burned out at this point and will probably just accept this one. Base is Menlo Park, CA (also looking for a female summer intern to split an apartment with!!)

Here's the interview report:

Round 1:
Product: About hashtags — it wasn't an A/B test design question. They asked how I'd roll out hashtag recommendations in the News Feed. It tested overall data analysis thinking, from data to modeling. The modeling part didn't seem to require any advanced recommendation models — they just asked about logistic regression.
SQL: I'd recommend grinding LeetCode's SQL 50. The actual interview question was pretty basic and wasn't one of the questions floating around the forum — it was about hashtags. Mainly watch your use of date functions and LEFT/RIGHT JOIN. After you finish writing, the interviewer asks some follow-up questions to test how well you actually understand SQL fundamentals — for example, the mechanics of different joins, why you'd use a LEFT JOIN, what a plain JOIN would give you instead, and what problems that could cause.

Round 2: The interviewer didn't seem super prepared either — they even asked me whether I was interviewing for full-time or for an internship. Maybe because my answers were on point, they didn't push with a lot of follow-ups, and the interview ended about 10 minutes early after just a few questions.
Product: About Facebook Dating — why it's good, and how you'd validate that. What the user trend would look like (I didn't draw anything, just described it verbally).
AE: I'm from a stats-adjacent major and had prepared some review material from back when I was doing quals, but I basically never used any of it. The question was roughly about network effects (spillover) — no calculation needed, just had to lay out the logic clearly.

Anyway, if you don't have SQL experience, I'd recommend grinding LeetCode's SQL 50. For product case prep I watched some YouTube videos — I'd recommend Amma Ding and Dan.

Wishing everyone good luck~

Published

Curated and edited by PracHub

Practice the questions from this interview

Discussion

Sign in to join the discussion. The author is notified of every comment.

Loading comments…

Interview at a glance

Company
Meta
Role
Data Scientist
Rounds
Technical Screen → Onsite
Outcome
Offer
Difficulty
hard
Interview date
Oct 2025
Questions from this interview
4 questions

Real Meta interview experiences

First-hand reports from Meta candidates — the rounds, the questions they were asked, and how it went.

All 125 Meta interview experiences