Find a lesson, question or project

Start typing to search published content and planned structures.

↑ ↓ moveEnter openEsc close

Google interview research

Candidate accounts · 2 scoped accounts · 6 reported questions/topics

Snapshot: 1 October 2026 · Software/backend focus · No live fetching

Self-reported accounts, not employer-confirmed. Preserve the stated role, team and location; this sample does not establish company-wide patterns or question frequency.

Questions people reported

Named problems, reported variants and broad topics are different kinds of evidence. Related practice links are not proof that the exact problem was asked. Some prompts were withheld; we do not invent their details.

Topics in these accounts: Coding

L4; Bengaluru, India

24 Aug 2026 · Publication date; interview date not confirmed

Read the original candidate account

This loop included trie, array/binary-search and graph problems, with complexity discussion, dry runs and a behavioral round.

Scope and source limitations

Keep this employer-specific role, date and region scope. Candidate reports are self-reported, not employer-confirmed.

This account was retained from the existing reviewed dataset and was not fully reread in the 1 October research pass.

  • Trie-based coding problem

    Coding · Reported topic

    The exact problem title was not given.

  • Array and binary-search problem

    Coding · Reported topic

    The round also included a graph/matrix follow-up.

  • Modified Dijkstra problem

    Coding · Reported topic

    The author described a hard graph problem, without complete constraints.

Interviewed for L4; reported L3 outcome; EMEA; city not stated

Apr 2026 · Interview date · Month only

Read the original candidate account

The author reported greedy/two-pointer, heap and DP interviews with follow-ups, plus Googleyness. Exact problem names were not supplied.

Scope and source limitations

Keep this employer-specific role, date and region scope. Candidate reports are self-reported, not employer-confirmed.

This account was retained from the existing reviewed dataset and was not fully reread in the 1 October research pass.

  • Two-pointer / greedy problem

    Coding · Reported topic

    A follow-up changed the space-efficiency requirement.

  • Heap problem with a follow-up

    Coding · Reported topic

    No exact question name was published.

  • Dynamic-programming problem

    Coding · Reported topic

    The author reported an unusual follow-up.

Additional independent, matching role/level/region reports are needed before claiming recurring patterns.

A suggested 14-day preparation plan

Authored practice, not the company’s syllabus. This general track is an editorial starting point, not a pattern inferred from all these reports. Confirm the format with your recruiter and adapt it to your actual level.

Algorithm reasoning

Prioritize invariants, traversal, hashing, heaps and changed-constraint follow-ups. Keep practical coding and design practice too.

  • Days 1–2: diagnose gaps with arrays/hashing and a graph exercise.
  • Days 3–5: traversal, dependencies, heaps and binary-search invariants.
  • Days 6–7: DP/backtracking; state, recurrence and edge-case tests.
  • Days 8–9: a timed coding mock with two follow-ups and a dry run.
  • Days 10–11: one executable design exercise and one service-design mock.
  • Days 12–14: project stories, a full mixed mock and targeted revision.
Practise an original design exercise

Design a document search service with pagination and ranking.

This exercise is written for practice, not reported by a candidate. Define requirements, implement the smallest useful version, then test concurrency, retries and failure recovery where relevant.

Customize a timed role/level prep brief

The timed brief uses a narrower, date/level-filtered dataset. Broader accounts on this page are deliberately not substituted into those filters.

How this research was collected

Public candidate accounts, official guides and original editorial practice are kept separate. We did not copy paid question banks or scrape all of the internet. Cross-posts are not independent evidence; one source can describe more than one company. Unknown or index-only dates are not used to claim recency. Future expansion needs permitted fetching, deduplication and metadata review; this version remains a static snapshot.