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...

Google SWE Intern Interview Guide: OA, Two DSA Rounds and What Was Asked

8 min read

Prepare for Google software engineer intern interviews with reported assessment patterns, interview questions, and preparation advice.

In this article
On this page

Last reviewed November 10, 2026.

For software engineering interns.

Built by the Codinzhub team from Google interview experiences that candidates shared, including the ones posted on Codinzhub. Each experience is one person's interview, and Google changes its process by year, level and team, so use this as a map, not a promise.

The 30-second version. A short recruiter call, an online assessment (usually 2 hard coding questions in 60 minutes), then two 45-minute DSA rounds with two questions each, coded in a Google Doc or Google's own editor with no compiler. Only candidates who pass round 1 get round 2. Some loops add a Googleyness round at the end. One 2027 prep session told candidates DP was not important; arrays, hashing, trees, graphs and heaps were.

Round-by-round breakdown

Stage 1: Application, where candidates applied

How it works. Most interns came through campus hiring (IIIT Bangalore, IIIT Allahabad and IIT BHU drives appear in these experiences); others used a referral or applied online. One intern was invited after finishing in the top 1,500 of a Google coding contest. Preparing for STEP instead? It has its own process: see the Google STEP guide.

What helped. Recruiters looked at coding profiles. They asked how many coding-practice problems the candidate had solved and checked Codeforces and CodeChef.

Stage 2: Shortlisting, how candidates moved forward

How it works. A 10 to 15 minute recruiter call: your introduction, projects, coding profiles and preferred language. One candidate said its main purpose was to check they were genuine. Several candidates then had a prep session that explained the interview format. One referred intern waited about 40 days before the next step.

Stage 3: Online assessment

How it works. Usually 2 coding questions in 60 minutes, with no time limit per question and, in one report, no proctoring. These are the hardest questions in the intern data. One 2025 OA had 15 multiple-choice questions plus 3 coding questions instead. Some interns had no OA and went straight to interviews.

At a glance

  • Length: 60 minutes, with no time limit per question
  • Questions per round: 2 coding questions (or 15 multiple-choice plus 3 coding)
  • What was asked: tree DP and re-rooting, range updates with DP, binary search with ordered sets
  • Most common: re-root a tree and XOR one subtree; the arithmetic-progression subsequence

OA questions candidates reported:

Question (our summary)What it tests
Re-root a tree and XOR one subtree with k to get the largest totalTree DP, re-rooting
Range additions on an infinite array, then the longest arithmetic-progression subsequenceSweep line, DP
Build the string for a tree node from its children's strings; check if it is a palindromeTree DFS, hashing
Longest all-zero subarray with at most X decrements and Y resetsBinary search + ordered set
Count students in three score bands; days to read a book chapter by chapterArrays, simulation

More questions: Browse Google questions.

Google OA questions shared on Codinzhub:

QuestionDriveTopic
Maximum Count Above Threshold in a WindowIIIT AllahabadSliding window
Ticket Queue ScoreIIIT AllahabadQueue simulation
Tier Task Count SumIIIT AllahabadSorting, binary search
Ways to Split ArrayOff-campusPrefix max and suffix min
Concord Interval CensusOff-campusSubarray max minus min
Lighthouse Flag BroadcastOff-campusStrings
The Runner-Up Relay RouteOff-campusSecond shortest path
Convoy Load Balancing With Escort WeightOff-campusBinary search on the answer
Resonant Carrier Frequency QueueOff-campusSliding window
Datacenter Node Priority AssignmentGoogle Big CodeGraphs, components
Mountain Range Peak OptimizerGoogle Big CodeDP
Recursive Cache Amplification MaximizerGoogle Big CodeDP
Tricky SubarraysNot statedDP on permutations
Sum of FV for All Nodes in a TreeNot statedTree DFS
Tasty CakesNot statedDP

Stage 4: The interview loop

How it works. Each round is 45 minutes: 5 to 10 minutes of introduction, 30 to 35 minutes for two DSA questions (often the second is a follow-up to the first), and a few minutes for your questions. You code in a Google Doc or Google's editor, which does not auto-close brackets. The interviewer sees your edits live. In the 2027 cycle, the last 5 to 10 minutes covered how you use AI tools at work. Only candidates who pass round 1 get round 2, sometimes 30 to 45 minutes later on the same day.

Round 1

At a glance

  • Length: 45 minutes: 5 to 10 minutes of introduction, 30 to 35 minutes of DSA, and in 2027 about 5 minutes on using AI at work
  • Questions per round: 2 questions, the second often a follow-up to the first
  • What was asked: graphs (ranks, union-find), trees, prefix sums and hashing, stacks
  • Most common: no single question repeats; Number of Visible People in a Queue and the friendship-log union-find are typical

Questions candidates reported:

Question (our summary)Follow-up reported
How many people the last person in a line can see; then how many everyone can see (Number of Visible People in a Queue)The second part decided the round
For each player, can their rank be fixed from match results?Do it in one DFS pass
Earliest time everyone is connected, from friendship logsHandle "no longer friends" logs online
Flip each leaf of a boolean expression tree: what does the root become?Skip subtrees that cannot change the answer
Largest sum between two equal values in an arrayFrom O(n²) to O(n) with a hash map
Group "buddy" strings whose adjacent-letter gaps matchNone
Score a string by adjacent-letter differences; return the best anagram from a dictionaryBuild it as a class; then a huge dictionary
First Missing Positive, after a 10-minute resume grillingO(1) space
A long config-merge problem; a knapsack that needs quickselectNot stated
Company hierarchy questions on an N-ary treeA harder follow-up
A medium BFS graph questionA tricky follow-up, discussed only

More questions: Browse Google questions.

Round 2

At a glance

  • Length: 45 minutes, sometimes 30 to 45 minutes after round 1 on the same day
  • Questions per round: 1 or 2 questions
  • What was asked: 2D prefix sums, multi-source BFS, iterator design and connected components
  • Most common: no single question repeats; the run-length iterator and the farthest-from-data-center BFS are typical

Questions candidates reported:

Question (our summary)What it tests
Smallest-sum submatrix of a fixed size2D prefix sums
The location farthest from any data centerMulti-source BFS
Run-length iterator with next and hasNext, then restart on a zeroIterator design
Pairs of people who do not know each otherConnected components
Pretty-print a binary treeLevel order traversal
A custom-sorted heap question; a binary-search-on-the-answer questionHeap, binary search
Maximum subarray when you may flip one numberArray DP

More questions: Browse Google questions.

Googleyness round (some loops)

At a glance

  • Length: not stated
  • Questions per round: not stated; 5 themes were reported
  • What was asked: your journey, pressure, teamwork, strengths and weaknesses, growth
  • Most common: none repeated

How it works. A behavioral round after the DSA rounds. Questions candidates reported: your journey so far, how you handle pressure, working in a team, strengths and weaknesses, and how you have grown.

Stage 5: Result

How it works. Results came 10 to 12 days after the last round in one report, and with the on-campus results in another. No report describes an internship converting to a full-time offer, so this guide makes no claim about that.

Google intern experiences shared on Codinzhub

Written up by students on Codinzhub. Open each one for the full rounds.

CollegeRole and resultWhat was asked
IIT BHUSDE intern, selectedWhen to use BFS and a level order traversal; Minimum Cost to Split an Array (partition DP), explained with pseudocode
IIT BHUSDE intern, not selectedEarliest time a process can message another as channels come online (union-find), in a 45-minute Google Doc round
IIIT AllahabadSDE intern, not selectedR1: the 3x3 subgrid with the largest sum. R2: every n up to 1000 that is a sum of two squares in two different ways
IIIT AllahabadIntern, not selectedA divide-and-conquer painting problem
ISM DhanbadSDE intern, not selectedR1: best subarray score minus its boundary elements. R2: the least time to finish tasks on two CPUs
Google Big Code routeSDE intern, not selectedThe kth smallest of all subarray sums (binary search, two pointers, prefix sums)
Not statedSWE intern, not selectedR1: count pairs with arr[i] - arr[j] = i - j, optimised with a hash map. R2: a tree BFS problem with tricky edge cases
Not statedSDE summer intern, selectedR1: the shortest of several queues when pop is expensive. R2: the nearest favourite city with Dijkstra
Not statedWeb solutions engineering intern, not selectedA 30 to 35 minute call on web basics and projects, then the shortest-queue problem

The Codinzhub Next-Round Playbook

What interns who moved to the next round did. Put together by the Codinzhub team from every Google intern experience in this guide.

To get shortlisted

  • Keep your coding profiles active. Recruiters asked how many problems you solved and looked at coding-practice, Codeforces and CodeChef.
  • Know every project on your resume. One interview opened with ten minutes of questions on past backend work.

To get past the OA

  • Expect two hard questions in 60 minutes, often on trees and range updates. Practise tree DP and re-rooting.
  • Secure partial marks: a correct slower solution beats an unfinished fast one.

To get past Round 1

  • Two questions in 30 to 35 minutes means about 15 minutes each. Watch the clock.
  • Give the brute force in the first two minutes, then improve it.
  • Finish the code for the second question. Interns who had the idea but not the code were rejected.
  • Dry-run on the interviewer's example before you say you are done.

To get past Round 2

  • Expect a different topic from Round 1, often graphs, 2D prefix sums or an iterator class.
  • Name variables clearly and add short comments. Interviewers said readability counts.
  • Aim to solve, dry-run and give the complexity within 20 to 25 minutes, as one IIIT Allahabad candidate advised.

To get past Googleyness

  • Prepare stories about pressure, teamwork, a weakness and how you have grown. One intern cleared every DSA round and was rejected here.

Topics to prioritise

Across these experiences, the most common topics are arrays and strings, trees, graphs, hashing, math and bits, binary search, DP, heaps. Trees and graphs come up in both the OA and the interviews. DP shows up mostly in the OA.

A four-week plan

  1. Week 1: Arrays, prefix sums and hashing. Subarray sums, first occurrence maps, 2D prefix sums.
  2. Week 2: Trees. DFS that returns a value, re-rooting, N-ary trees, level order.
  3. Week 3: Graphs and heaps. BFS, multi-source BFS, union-find, custom comparators.
  4. Week 4: Timed mocks. Two questions in 35 minutes in a plain document, out loud, with a dry run.

FAQ

Is DP needed for a Google internship?

It appears in OAs, but one 2027 prep session said it was not important for the interviews. Cover the basics and spend most of your time on trees, graphs and arrays.

What about STEP?

STEP is a separate program for first- and second-year students, with its own timeline. See the Google STEP guide.

About this guide

This guide is researched and written by the Codinzhub team. We read every Google interview experience we could find, kept only what candidates actually described, and rewrote it in our own words. Questions are our own short summaries; the practice links point to similar problems, not the exact interview questions. 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.