Candidates describe Meta's Software Engineer loop as a recruiter conversation, a coding screen, then a loop of coding, system design and behavioral rounds. Recent reports add an AI-enabled coding round built around debugging and testing.
$206K
Avg. Base Comp
$430K
Avg. Total Comp
4-6 rounds
Typical Rounds
Not reported
Process Length
Candidates describe Meta's Software Engineer interview as a coding-first process that is changing. Most reports start with a recruiter conversation and a LeetCode-style technical screen, then a loop that combines coding, system design and a behavioral round. Reported loop sizes vary, from four rounds to a longer process with an online assessment and onsite stage.
The newest reports are the most distinctive. Two candidates describe an AI-enabled coding round. One was given a small project with a bug, had to write test cases, and could use an embedded LLM at any point. The other says interviewers judged how they validated AI-generated ideas and debugged or refined the output. Another candidate described a multi-file online assessment that required navigating existing abstractions across a larger codebase. Practice reading unfamiliar code, not only solving isolated puzzles.
Classic algorithm questions still appear. Candidates mention BFS on a maze, lowest common ancestor, longest consecutive sequence, Alien Dictionary, Kth Largest Element, graph and tree problems, and medium-level array and string questions. Several stress speed: starting almost immediately, moving from brute force to an optimal solution, explaining complexity and doing a dry run. One experienced candidate who was rejected said interviewers kept pushing on why each step was chosen, even after correct answers.
System design reports include ad click aggregation, a filtering system for tracking a type of post or ad, and scalable API design for a high-traffic application. Behavioral rounds are described as STAR-based, with project deep dives, conflict resolution and teamwork questions. Most shared outcomes in this set were rejections, so treat the loop as demanding.
Synthesized from 169 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Meta process.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Meta
Select the 2nd highest salary in the engineering department
| Question | |
|---|---|
| Employee Salaries | |
| Empty Neighborhoods | |
| Subscription Overlap | |
| Merge Sorted Lists | |
| Comments Histogram | |
| Like Tracker | |
| Random SQL Sample | |
| Weighted Keys | |
| Flight Records | |
| Largest Salary by Department | |
| Top 3 Users | |
| Find the Missing Number | |
| Recurring Character | |
| Scrambled Tickets | |
| Liked Pages | |
| Find the First Non-Repeating Character in a String | |
| Twenty Variants | |
| Find Bigrams | |
| Session Difference | |
| Detecting ECG Tachycardia Runs | |
| One Element Removed | |
| Size of Joins | |
| Last Transaction | |
| Threaded Comments | |
| Friendship Timeline | |
| The Brackets Problem | |
| P-value to a Layman | |
| Level Of Rain Water In 2D Terrain | |
| Get Top N Frequent Words |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report an initial recruiter call about their background, the role and expectations. One candidate notes a recruiter asked about their notice period early on. This is typically a qualification and logistics step before the technical rounds begin.
Candidates report a phone or video screen with LeetCode-style questions, such as binary tree, sliding window, palindrome and array problems. One candidate describes it as 45 minutes. Some mention explaining complexity, doing a dry run and covering edge cases aloud. One report also describes a multi-file online assessment, so the format may vary.
Loop reports describe medium problems and occasionally harder ones across graphs, trees, arrays, dynamic programming and linked lists. Candidates say the pace is fast, with little small talk, and that interviewers may push for the most optimal solution and ask why each choice was made. Practicing timed problems and narrating tradeoffs is likely to help.
Two candidates report a newer format. One got a small project with a bug, wrote test cases and could use an embedded LLM. The other says interviewers evaluated how they validated and refined AI-generated ideas. Practicing testing and debugging code you did not write may help.
Candidates report design prompts like ad click aggregation, post or ad filtering and high-traffic API design. Behavioral rounds are described as STAR-style, covering project impact, ownership, conflict and teamwork. One candidate mentions a team-matching step after the loop.
