Google Software Engineer L3 Interview Questions (2025)

What Is Google L3?

Google L3 is the standard entry-level Software Engineer role for new-grad and early-career hires. It's equivalent to an SWE I at most other companies and is the most common offer target for university recruits and candidates with 0–2 years of experience.

Despite being "entry-level" at Google, L3 interviews are notoriously difficult — you're competing with candidates from top CS programs worldwide and expected to solve Medium–Hard LeetCode problems under time pressure.

The L3 Interview Loop

A typical Google L3 loop (virtual or onsite) consists of 4–5 rounds:

Note: Some L3 loops skip system design entirely. Whether it's included depends on the team and your experience level.

Real Google L3 Coding Questions (Reported 2023–2025)

These questions have been reported by candidates who went through the L3 loop. They are grouped by topic, as Google tends to focus on specific pattern areas.

Arrays & Hashing

Find two numbers that add up to a target. Follow-ups: what if there are duplicates? What if the array is sorted?

Given a list of strings, group those that are anagrams of each other.

Follow-up: optimise from O(n·k log k) to O(n·k) using character frequency as the key.

Find the longest sequence of consecutive integers in an unsorted array in O(n) time.

Return an array where each element is the product of all other elements — without using division.

Strings & Sliding Window

Find the smallest window in string S that contains all characters of string T.

This is a Google favourite. Master the two-pointer + frequency map pattern.

Classic sliding window. Be ready to extend it — "at most K distinct characters".

Stack-based problems. Interviewers often follow up with "generate all valid combinations of N pairs".

Trees & Graphs

BFS on a tree. Follow-ups include zigzag traversal, right-side view, and average of levels.

Both variants appear. Know the recursive DFS approach for binary trees and the BST shortcut.

BFS shortest path on an implicit graph. This is a real L3 question seen at Google. Tricky to implement cleanly under pressure.

BFS or DFS matrix traversal. Often extended to "number of enclaves" or "surrounded regions".

Detect cycles in a directed graph using Kahn's algorithm or DFS-based cycle detection.

Dynamic Programming

Classic DP. Be ready to explain the recurrence relation and space optimisation.

Two-string DP. Follow-up: what's the minimum number of insertions/deletions to make them equal?

Greedy and DP approaches. Google interviewers often ask for both and compare trade-offs.

Count the number of ways to decode a digit string. Classic medium DP with edge cases.

Heaps & Priority Queues

Min-heap of size K. One of the most common Hard problems at Google.

Bucket sort or min-heap. Know both approaches and their trade-offs.

Two heaps (max-heap + min-heap). Appears more at L4+ but has been reported at L3.

Design & OOP (Less Common at L3)

HashMap + doubly-linked list. The canonical design question at Google L3/L4.

Object-oriented modelling. Tests whether you can structure classes, interfaces, and relationships cleanly.

Google L3 Behavioural Questions

The "Googleyness" round evaluates cognitive ability, collaboration, and integrity. Common questions:

• *"Tell me about a time you had a disagreement with a team member. How did you resolve it?"*
• *"Describe a project you're most proud of. What was your specific contribution?"*
• *"Give me an example of when you had to learn something new quickly."*
• *"Tell me about a time you made a mistake. What did you do?"*
• *"Describe a situation where you had to work with ambiguous requirements."*

Situation — set the scene

Task — what you were responsible for

Action — exactly what *you* did (not "we")

Result — measurable impact

Google Differentiator: Google specifically values intellectual humility. Answers that show you learned from failure score higher than heroic "I saved the day" stories.

L3 System Design (If It Comes Up)

For L3, system design expectations are lower than L4/L5 — but you should still be able to:

• Clarify requirements and define scale (users/QPS/storage)
• Sketch a simple high-level architecture (client, API server, database)
• Explain basic caching and when you'd use it
• Discuss database choice (SQL vs NoSQL) with rationale
• Identify one or two bottlenecks and how to address them

• Design a URL shortener
• Design a key-value store
• Design a notification service
• Design a rate limiter

Google L3 Preparation Timeline

Aim for 150–200 LeetCode problems before your interview, with a strong focus on Mediums.

Interview-Day Execution Framework

• Listen fully before repeating the problem in your own words
• Ask 2–3 clarifying questions (input constraints, edge cases, return type)
• State your approach and complexity before typing a single line
• Code with narration — explain what each section does
• Test your code with the given example, then an edge case (empty input, single element)
• Optimise only if asked — a clean correct solution beats a clever broken one

How Topalupu Helps You Crack L3

Topalupu's Google prep track includes:

• Problems matched to the exact L3 question bank above
• AI mock coding interviews with a Google engineer persona and real-time feedback
• Behavioural sessions scored using the STAR framework
• System design walkthroughs targeted at L3 expectations

Every session is timed, realistic, and designed to build the muscle memory you need on interview day.