Jobra
🐞Step 1 of Jobra:AI Debugging Rounds
DSA+AIPreparing you for both.
JOBRA · AI DEBUG ROUND (PREVIEW)sum.cpp
// buggy snippet
int total(vector<int>& a, int n) {
    int sum = 0;
    for (int i = 0; i <= n; i++) { // 🐞        sum += a[i];
    }
    return sum;
}
A real codebase. A real bug. You + AI.

Jobra drops you into a broken repo. Find it, fix it, ship it. AI helps you debug like on the job.

Prompt:Correct the loop condition
Tap 'Let Jobra fix it' below to simulate AI debugging...

Meta Data Engineer Interview Guide: Timed SQL and Python, Product Sense and Data Modeling

3 min read

Prepare for Meta data engineer interviews with reported assessment patterns, interview questions, and preparation advice.

In this article
On this page

Last reviewed November 15, 2026.

For data engineer candidates.

Built by the Codinzhub team from Meta data engineer interview experiences that candidates shared. Each experience is one person's interview, so use this as a map, not a promise.

The 30-second version. A recruiter call, then a technical screen of 25 minutes of SQL and 25 minutes of Python (one recruiter said you need about three of each right). The onsite has three product rounds and one behavioral round. Each product round uses one scenario (Reels, car rental, carpool, private accounts) and moves fast through product sense, data modeling, SQL and Python. What you write down decides the result, so time management is everything.

Round-by-round breakdown

Stage 1: Application, where candidates applied

How it works. Most candidates were contacted by a recruiter, sometimes several times. One senior candidate applied directly.

Stage 2: Shortlisting, how candidates moved forward

How it works. A recruiter call about fit and the process. HR explained each onsite round in advance.

Stage 3: Technical screen

At a glance

  • Length: about 50 minutes (25 SQL, 25 Python)
  • Questions per round: about 3 SQL and 3 Python questions
  • What was asked: SQL on a library schema, Python with dictionaries and heaps
  • Most common: the library tables (books, copies, checkouts, members)

Questions candidates reported:

Question (our summary)What it tests
Most points from up to 3 books, each from a different categoryPython, greedy
Books in good condition never returned; top 3 by longest loan among books with more than 10 copiesSQL
Reassign people when office buildings close, using a capacity tablePython, dictionaries
Values only the current user hasPython, sets

More questions: Browse Meta questions.

Stage 4: The onsite

How it works. Four rounds, often on one day. Three product rounds, each about one scenario, in this order: product sense (about 10 minutes), data modeling (about 15), SQL (about 15) and Python (about 15). One candidate's three scenarios were Reels, car rental and carpool. The last round is behavioral and ownership.

Product rounds

At a glance

  • Length: about an hour each
  • Questions per round: 4 to 6 metrics, a small data model, 2 to 3 SQL and 1 to 2 Python questions
  • What was asked: success metrics, fact and dimension tables, cumulative metrics, sweep lines
  • Most common: carpool; Netflix-style streaming; private accounts

Questions candidates reported:

Question (our summary)What it tests
Count active follows from request, accept, reject and unfollow eventsSQL with event order
Recommend people followed by the people a user followsPython, dictionaries
Can every shared ride be completed with the car's seats?Python, sweep line
Cars rented and inventory per city and car size at the end of a daySQL joins
A cumulative metric where today's IDs may be missing yesterdaySQL
Top 3 rated items per categoryPython

Behavioral round

At a glance

  • Length: about 45 minutes
  • What was asked: projects from 0 to 1, conflict, priorities, persuading others
  • Most common: a conflict and a deadline story

More questions: Browse Meta questions.

Stage 5: Decision

How it works. Mixed results led to an add-on interview for one candidate. One senior candidate was rejected for not showing enough project impact.

The Codinzhub Next-Round Playbook

What data engineers who moved on did. Put together by the Codinzhub team from the DE experiences in this guide.

To get past the screen

  • Practise short Python problems until each takes under 7 minutes.
  • Write SQL in a single scan where you can, and clarify exact matches.

To get past the onsite

  • Keep product sense short: 4 to 6 metrics you can reuse in data modeling.
  • Read chapters 1 to 3 of a data warehouse toolkit for facts, dimensions and slowly changing dimensions.
  • Ask the interviewer to help track time for each section, and write everything down.

About this guide

This guide is researched and written by the Codinzhub team. We read every Meta data engineer interview experience we could find, kept only what candidates actually described, and rewrote it in our own words. Questions are our own short summaries. Found something out of date? Share your own experience on Codinzhub and we will update the guide.

Ready for the next topic?

Keep track of what you’ve learned.