How Google interviews
These are the Google loops we have researched. Each one lists the rounds, what they test, the follow-up questions to expect, and how to prepare, with the sources we used.
- Software Engineer
- Mid-level · 5 rounds · 3 h 45 min
- Product Manager
- Mid-level · 4 rounds · 2 h 40 min
- UX Designer
- Mid-level · 4 rounds · 2 h 40 min
What Google says about its interviews
Google's public material does not establish a fixed current PM loop order or per-round duration, so 40-minute practice rounds are representative fallbacks. Google Careers: interview tips
Google's official portfolio guidance asks for three to five case studies showing end-to-end process, role, impact, collaboration, and design rationale. Google UX design portfolio tips
Google's virtual interview guide explicitly anticipates diagramming for some UX interviews. Google virtual interviews candidate guide
Google Software Engineer interview
Prepare for a coding screen, three algorithmic interviews, and a Googleyness and leadership conversation. L4 coding rewards turning an unfamiliar problem into a precise model, implementing it clearly, and adapting to follow-ups. Prioritize graphs, intervals, trees, and stateful data structures; retain breadth in recursion and mathematical reasoning. Practice complete solutions with complexity analysis and manual traces, alongside concrete collaboration and ambiguity stories.
The 5 rounds
Technical coding screen
Establish independent problem solving: turn a verbal specification into correct code and explain why it works.
Expect: A traversal or interval problem with a changed input constraint.
Follow-ups: “What breaks with duplicates or disconnected input?” “How does the solution change when the input becomes a stream?”
Coding: trees and graph state
Expose whether you can represent a graph problem precisely and retain correctness under a harder traversal constraint.
Expect: A graph or tree traversal extended with path constraints.
Follow-ups: “Can the same vertex be revisited under a different state?” “What changes with weighted edges?”
Coding: sequences and reasoning
Demonstrate reasoning beyond pattern recognition through a clear state model, implementation and proof by cases.
Expect: An interval or sequence problem requiring a justified greedy or dynamic state model.
Follow-ups: “Which counterexample defeats the greedy alternative?” “Can the same state answer an additional query?”
Coding: stateful data structures
Show that an API, its internal state and its per-operation costs remain consistent as requirements expand.
Expect: A stateful collection with efficient updates, queries and an added operation.
Follow-ups: “How would snapshots or deletion change the representation?” “What is the memory cost after many updates?”
Googleyness and leadership
Assess how you make decisions, collaborate through disagreement and learn while taking responsibility for outcomes.
Expect: A disagreement that changed your approach. A hypothetical ambiguity or collaboration problem. A failure followed by a concrete improvement.
Follow-ups: “What did the other person believe?” “What would make you change your decision?”
What candidates and sources report
- L4 candidates describe a technical screen followed by three coding interviews and a Googleyness conversation. Candidate report
- Graph traversal appears in both screening and onsite reports, with changes to constraints after the initial solution. Candidate report
- A May 2025 L4 candidate describes class design with additional operations and harder data-structure follow-ups. Candidate report
- An August 2026 L4 report includes trie search, array/binary search, and a modified Dijkstra problem. Candidate report
- A March 2026 L4 candidate describes a medium-hard tree round and a separate mathematical-reasoning problem. Candidate report
How to prepare
- Establish a coding baseline: Do one unseen 45-minute round. Record where modeling, implementation or verification consumed the clock; drill the weakest step first.
- Master traversal state: Rotate graph and tree problems. Define visited state explicitly, then add a constraint that changes what a node means.
- Broaden sequence reasoning: Practice intervals, binary search and recurrence derivation. Give a counterexample before rejecting an alternative.
- Build an evolving API: Implement a small collection, then add snapshots or deletion. Trace an interleaved sequence and state each operation’s cost.
- Prepare decisions, not scripts: Build six short stories on conflict, feedback, failure and ambiguity. Rehearse skeptical follow-ups and one hypothetical variant.
Mistakes to avoid
- Do not recite a remembered solution without checking constraints.
- Do not leave the final complexity implicit.
- Do not mark a state visited before defining what makes it unique.
- Do not assume every graph is a tree.
- Do not force dynamic programming onto a simpler problem.
Questions to ask your interviewers
- How does this team distinguish strong L4 ownership from L5 scope?
- Which user or reliability constraint drives the team’s hardest decisions?
- How do engineers influence work beyond their immediate project?
Google Product Manager interview
Prepare for product vision and design, analytical judgment, execution under technical constraints, and Googleyness and leadership. Current firsthand accounts agree on these signal families but do not establish one universal stage order or duration. At the mid-level bar, drive an ambiguous problem from user choice to measurable decision, expose second-order effects, and make engineering and stakeholder trade-offs explicit.
The 4 rounds
Product vision and design
Learn whether the candidate can choose a user, frame an important problem, set a direction, and defend a coherent product concept.
Expect: An ambiguous product vision problem spanning immediate value and a multi-year direction.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
Analytical and metrics judgment
Learn whether the candidate can define success, diagnose a metric movement, and choose an evidence-backed action under incomplete information.
Expect: A product metric diagnosis with segmentation, hypotheses, and decision thresholds.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
Technical product execution
Learn whether the candidate can turn product intent into a shippable plan while reasoning credibly with engineering about constraints, risk, and scope.
Expect: A roadmap and launch decision constrained by latency, privacy, reliability, or platform dependencies.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
Googleyness and leadership
Learn how the candidate leads through ambiguity, seeks perspectives, handles disagreement, and updates decisions after learning that an assumption was wrong.
Expect: Past examples of influence without authority, ambiguity, feedback, and inclusive decision making.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
What candidates and sources report
- A current mid-level Google PM posting emphasizes user journeys, strategy, cross-functional delivery, launches, experiments, and success metrics. Job posting
- A recent former-Google account describes product design, analytical thinking, technical depth, and behavioral or leadership as core PM interview components. Candidate report
- Mid-level preparation should connect user choice, strategic direction, measurable outcomes, and feasible delivery rather than treating them as isolated frameworks. Job posting
- Google's public material does not establish a fixed current PM loop order or per-round duration, so 40-minute practice rounds are representative fallbacks. Google
How to prepare
- Practice product vision and design: Run one 40-minute rep using an ambiguous product vision problem spanning immediate value and a multi-year direction. Use “Frame the user problem” and “Defend the strategy” as checkpoints; repeat after one changed constraint.
- Practice analytical and metrics judgment: Run one 40-minute rep using a product metric diagnosis with segmentation, hypotheses, and decision thresholds. Use “Define the metric precisely” and “Make a measured decision” as checkpoints; repeat after one changed constraint.
- Practice technical product execution: Run one 40-minute rep using a roadmap and launch decision constrained by latency, privacy, reliability, or platform dependencies. Use “Break down delivery risk” and “Align cross-functional partners” as checkpoints; repeat after one changed constraint.
- Practice googleyness and leadership: Run one 40-minute rep using past examples of influence without authority, ambiguity, feedback, and inclusive decision making. Use “Own an ambiguous outcome” and “Update after learning” as checkpoints; repeat after one changed constraint.
Mistakes to avoid
- Do not substitute a memorized framework for the supplied facts.
- Do not hide personal decisions behind vague team language.
- Do not claim certainty beyond the evidence available.
Questions to ask your interviewers
- Which PM signal is the hiring team most concerned about for this role?
- How does this team divide product strategy and execution ownership between PM, UX, and engineering?
- What distinguishes strong mid-level scope from senior PM scope on this team?
Google UX Designer interview
Prepare a concise portfolio narrative, a live end-to-end design challenge, craft critique, and collaboration examples. Google’s current mid-level interaction-design postings emphasize user-centered methods, flows, prototypes, design systems, experimentation, and cross-functional work. Candidate evidence supports portfolio and challenge formats, but much of the detailed public reporting is internship-level, so confirm the exact mid-level schedule with the recruiter.
The 4 rounds
Portfolio narrative
Learn whether the candidate can explain an end-to-end design decision, their own contribution, collaboration, iteration, and measured outcome.
Expect: A verbal portfolio walkthrough of one complex shipped project and one sharply different project.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
Live product design challenge
Learn whether the candidate can explore an ambiguous UX problem, map a coherent flow, test assumptions, and iterate under critique.
Expect: A whiteboard challenge requiring user choice, journey mapping, key states, and a changed constraint.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
Interaction craft and critique
Learn whether the candidate can diagnose an interaction, apply hierarchy and system patterns, and defend detailed choices across states and accessibility needs.
Expect: A product-flow critique covering hierarchy, interaction states, accessibility, and design-system consistency.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
Design collaboration and influence
Learn how the candidate uses research, handles disagreement, raises the craft bar, and gets quality work shipped through cross-functional constraints.
Expect: Past examples of research-driven change, engineering negotiation, critique, and design-system stewardship.
Follow-ups: “Change one important constraint and ask what survives.” “Probe the weakest assumption and the evidence needed.”
What candidates and sources report
- Google's official portfolio guidance asks for three to five case studies showing end-to-end process, role, impact, collaboration, and design rationale. Google
- A current mid-level Interaction Designer posting requires a portfolio and emphasizes user-centered design, cross-functional collaboration, prototypes, and design systems. Job posting
- Google's virtual interview guide explicitly anticipates diagramming for some UX interviews. Google
- A defensible mid-level practice loop should evaluate portfolio storytelling, live design reasoning, craft critique, and collaboration without assuming internship timing transfers exactly. Google
- The strongest practice bar is a traceable chain from user problem through alternatives and iteration to shipped impact, with the candidate's own contribution explicit. Google
How to prepare
- Practice portfolio narrative: Run one 40-minute rep using a verbal portfolio walkthrough of one complex shipped project and one sharply different project. Use “Frame the design problem” and “Demonstrate design impact” as checkpoints; repeat after one changed constraint.
- Practice live product design challenge: Run one 40-minute rep using a whiteboard challenge requiring user choice, journey mapping, key states, and a changed constraint. Use “Choose the design scope” and “Evolve the flow” as checkpoints; repeat after one changed constraint.
- Practice interaction craft and critique: Run one 40-minute rep using a product-flow critique covering hierarchy, interaction states, accessibility, and design-system consistency. Use “Diagnose the interaction” and “Give actionable critique” as checkpoints; repeat after one changed constraint.
- Practice design collaboration and influence: Run one 40-minute rep using past examples of research-driven change, engineering negotiation, critique, and design-system stewardship. Use “Navigate design disagreement” and “Learn from compromise” as checkpoints; repeat after one changed constraint.
Mistakes to avoid
- Do not substitute a memorized framework for the supplied facts.
- Do not hide personal decisions behind vague team language.
- Do not claim certainty beyond the evidence available.
Questions to ask your interviewers
- Which project should the portfolio walkthrough prioritize for this team?
- Will the loop include live diagramming, critique, or another functional exercise?
- How does this team evaluate mid-level craft versus design leadership?
Sources
Research reviewed . Practice exercises are our own; they are not confidential interview questions.
- Google Careers: interview preparation — Google
- Google L4: phone screen and onsite, April 2025 — Candidate report
- Google L4 interview experience, May 2025 — Candidate report
- Google L4 interview experience, August 2026 — Candidate report
- Google L4 onsite: trees and mathematical reasoning, March 2026 — Candidate report
- Google L4 offer: screen, three coding rounds and Googleyness — Candidate report
- Google Careers: Product Manager, Googler Experience — Job posting
- Google PM interview experience discussion — Candidate report
- Google interview questions: Product Manager card, March 11, 2026 — Candidate report
- Google UX design portfolio tips — Google
- Google Careers: Interaction Designer, Sales Platform — Job posting
- Google UX Design intern interview expectations — Candidate report
- Google virtual interviews candidate guide — Google
Independent preparation using Google as a practice target. Not affiliated with, endorsed by, or sponsored by Google.