RepostedGoogle logo

Google · Onsite

In process

Google Intern Knapsack and Trees: 17 Experiences (Part 6 of 8)

Software engineer · InternshipSep 2026
Shared byjanebeingjane

Google SWE intern reports from Sep to Oct 2026: two coding rounds with tree merging, knapsack and 2D DP, then team matching, waits and a few rejections.

At a glance

  • Process: two back-to-back technical interviews, mostly LeetCode variations with no behavioral questions; results came from the next morning to about a week later, usually as a move to Project Search (59, 60, 61, 70).
  • Asked recently: knapsack variants in 2 of 17 reports (66, 72); tree merging (66), a monotonic stack (60), 2D DP (72), and an oral scenario plus 2D prefix sums and simple API design (62) once each.
  • Difficulty: candidates called it medium (70) to medium-hard (59), harder than earlier years (72), and very dependent on the interviewer (61, 66). Long spoken prompts used up time (62, 66).
  • Practise first: Merge Hierarchical Trees by Name, Unit-Weight Knapsack and Maximum Task Reward Within a Time Budget.

Coverage: Google Summer 2027 SWE and SWE PhD intern loops, plus several commenters whose role or cohort was not stated. One report from North America; others unspecified. Seventeen reports, entries 57–73: fifteen from September–October 2026 and two looking back at 2025 interviews.

In this series

  • Part 1: graphs, intervals and OOD from the first wave of 2027 intern loops.
  • Part 2: paths, grids and trees.
  • Part 3: prefix sums and heaps.
  • Part 4: full loops and team matching.
  • Part 5: grids, DP, OOP and AI.
  • Part 6 (current): trees and knapsack.
  • Part 7: PhD and return offers.
  • Part 8: updates and host calls.

What came up

Most of these accounts are about what happened after the interviews. The few that named topics point to DP, with two different knapsack variants.

Topic Reports Last seen
Knapsack 2 of 17 Oct 2026 (report date)
Two-dimensional DP 1 of 17 Oct 2026 (report date)
Tree merging, DFS and hashing 1 of 17 Oct 2026 (report date)
Greedy and Quickselect 1 of 17 Oct 2026 (report date)
Monotonic stack 1 of 17 Oct 2026
Scenario-wrapped algorithm 1 of 17 Oct 2026 (report date)
2D prefix sums 1 of 17 Oct 2026 (report date)
Simple API design 1 of 17 Oct 2026 (report date)

A candidate can appear in more than one row.

Questions we found

Algorithms / Data Structures

Question Reports Last seen
Merge Hierarchical Trees by Name 1 Oct 2026 (report date)
Unit-weight knapsack with limited quantities 1 Oct 2026 (report date)
0/1 knapsack-like 2D DP (practice) 1 Oct 2026 (report date)

17 detailed candidate timelines, loops, and screens

65: A rejection and an honest look at prep

Report date: Oct 20, 2026 · Role: Summer 2027 Software Engineering Intern · Result: Rejected

  1. An OA, then an interview, then a rejection email. Their own questions, rounds and dates were not shared.
  2. Reading another candidate's account, they named their gaps: prerequisite checks in recursion follow-ups, rusty DP and BFS, weak graphs, and explaining complexity aloud while nervous.

64: Two rounds on October 19

Interview date: Oct 19, 2026 · Role: not stated

  1. Two rounds done; they asked how long results take and whether a quick result meant a competing offer. No questions or decision were shared.

73: A 30-minute second round

Report date: Oct 14, 2026 · Role: not stated · Status: Awaiting a decision

  1. Their second round ended after 30 minutes, with no reason given. The result had not arrived in a later reply.

72: Two rounds of 2D DP

Report date: Oct 14, 2026 · Role: Summer 2027 SWE Intern · Location: North America · Stage: Virtual onsite

  1. Two interviewers, two 2D DP questions. They had not reviewed DP because other companies never asked it, and felt this year was harder than before, partly by luck.
  2. The first resembled 0/1 knapsack and they wrote it from memory (practice; the exact variation was not shared).
  3. They finished an answer in round two without knowing if it was right; no confirmation and no time for a follow-up. They expected a rejection.

60: Team matching the next morning

Interview date: Oct 14, 2026 · Role: Summer 2027 SWE PhD Intern · Status: Project Search / team matching

  1. Applied Sep 30; two back-to-back technicals Oct 14.
  2. Round 1: a monotonic-stack task with a friendly interviewer. Round 2: three questions, each followed by complexity and "can this be better?" None of the prompts were shared.
  3. Moved to Project Search the next morning. The email said matching was rolling from October with no guarantee, the questionnaire would come within 24–48 hours, and deadlines should be flagged. They filled it in and were waiting.

66: Tree merging, unit-weight knapsack, then a rejection

Report date: Oct 13, 2026 (outcome Oct 15) · Role: Summer 2027 SWE PhD Intern · Result: Rejected

  1. Round 1: a friendly interviewer read out a tree-merging task with five detailed rules, part of it pasted into a doc. Nodes have string names, integer values and ordered children; matched nodes keep the first tree's names and order and the second tree's values; children pair by name in order of occurrence. About 20 minutes went to understanding it. They chose DFS and wrote the base case and two loops but not the matching logic, and later suggested DFS plus hashing for duplicates. Simpler to code than LeetCode 1932, with a prompt two to three times longer.
  2. Round 2: a unit-weight knapsack: every item weighs one unit, quantities vary. A greedy idea came fast, but clarifications through an accent they found hard, plus about ten minutes on a bug, left three minutes to optimize. They discussed DFS with memoization at O(n × W); later they pointed to an O(n) Quickselect direction they had not prepared.
  3. A recruiter rejection came Oct 15 with no detailed feedback; they put it down to not finishing round one.

61: Still "Interview scheduled," then team matching

Report date: Oct 12, 2026 · Role: Google 2027 SWE Intern · Status: Team matching

  1. Before interviewing they found few 2027 reports. Afterward their careers page still said "Interview scheduled" and they had no referrer to check; a later edit confirmed team matching.
  2. Mostly LeetCode variations, no system design. They felt the question pool was large and format and difficulty varied a lot by interviewer.

62: An oral scenario, prefix sums and an API question

Report date: Oct 12, 2026 · Role: not stated · Status: Awaiting a decision

  1. An algorithm wrapped in an application scenario, explained entirely aloud. They spent a long time pinning down what was wanted, wrote interfaces and logic in a doc, and checked their understanding; once clear, it was manageable. It was not OOP.
  2. Then a 2D prefix-sum question and a very simple API design question.
  3. Interviewed on a Monday; still "pending review", which they understood as waiting for a hiring committee, with the careers page showing "Submitted".

71: Waiting for the questionnaire

Report date: Oct 2026 · Role: not stated · Status: Waiting for the project questionnaire

  1. They had the good-news message but no project questionnaire yet, and asked whether the two usually arrive together.

68: A non-CS applicant's surprise invite

Report date: Oct 2026 · Role: not stated · Result: Rejected

  1. Without a CS background, they applied on encouragement, were surprised by an interview invite, and started practicing coding. The application ended in rejection; they plan to apply again.

63: Can a PhD intern do research?

Report date: Oct 2026 · Role: PhD SWE Intern · Status: Waiting for team matching

  1. They asked two interviewers whether a PhD SWE intern could do research and publish. Both said yes in principle, but most teams focus on product work.

70: A canceled interview, a makeup round, then team matching

Interview date: Sep 22 and 25, 2026 · Report date: Oct 5, 2026 · Role: Summer 2027 PhD SDE Intern · Status: Team matching

  1. Applied without a referral Aug 24; scheduling Sep 3. Two back-to-back interviews were set for Sep 22; one was canceled at short notice and made up Sep 25. Team matching Oct 1.
  2. Both medium, LeetCode-style variations that the interviewers kept building on. No resume deep dive; graduation timing and transcript were asked at scheduling.
  3. The Project Search email: last stage before an offer, rolling matching from October, questionnaire in 24–48 hours, possible host interviews, an offer only on a mutual match.

57: A week of waiting after an unexpected question

Report date: Oct 4, 2026 (update Oct 6) · Role: not stated

  1. An unexpected kind of question, not described. They heard back exactly one week later but did not share the decision.

58: A reasoning question, then four easy coding questions

Report date: Sep 29, 2026 · Role: not stated

  1. Round 1: one pure reasoning question, not a LeetCode problem; the feedback seemed okay.
  2. Round 2: four easy LeetCode questions, which surprised them. None of the five were named.

59: An unusual second round, then team matching

Report date: Sep 25, 2026 (update Oct 6) · Role: Google 2027 SWE Intern · Status: Team matching

  1. Round 1 went very well. Round 2 was unusual enough that routine practice would not have helped; with hints they solved it. They later called it medium-hard.
  2. The interview invite came the day after passing the OA; the OA had no camera; the technicals had no behavioral questions. Team matching was confirmed Oct 6.

67: A kind interviewer and a question they could not code

Interview date: 2025 · Role: SWE Intern

  1. A non-CS candidate got a simple question from a kind interviewer but could not code it, and chose not to apply for SWE again. They suggested the PhD SWE intern route to other non-CS candidates.

69: Recruiters came back the next year

Interview date: 2025 (follow-up 2026) · Role: not stated

  1. One interview went poorly, another okay. HR contacted them again in 2026. Their 2027 PhD intern account is in Part 7.

Patterns across the reports

  1. Quick handoff to Project Search. The next morning (60) and six days after the makeup round (70); 59 confirmed it in an edit 11 days after interviewing.
  2. Long prompts eat the clock. Spoken or long prompts took 20 minutes or more to understand in 62 and 66.
  3. No behavioral round. Three candidates said the technicals had no behavioral or resume questions (59, 61, 70).

Common questions

How fast does Google reply after intern interviews?

From the next morning (60) to six days (70) or a week (57). Two others were still waiting (62, 73).

What does "pending review" mean in a Google intern application?

One candidate read it as waiting for a hiring-committee review; their careers page still showed "Submitted" (62).

Can a Google PhD SWE intern do research?

Two interviewers told one candidate it is possible in principle, but most teams do product work (63).

Keep practicing

In this series

Sources: Anywhere useful interview info can be found.

0 total reactions. Your reactions: none.
0 comments

Comments

0

No comments yet — be the first to share what you know.