L4 Software Engineer Interview Prep
Curated practice for the Google SWE interview loop, calibrated to L4 expectations.
About this loop
Google's L4 SWE loop is built around two coding rounds, a system design round, and a Googleyness/leadership round. Coding rounds emphasize algorithmic clarity, edge-case handling, and clean code over raw speed. System design enters the loop at L4, but expectations are bounded - you should be fluent in fundamentals (caching, sharding, load balancing, queues) without needing to design a full Spanner-scale system. The Googleyness round is not a soft round - structured behavioral questions probe collaboration, ambiguity, and your motivation for working at Google. This pack groups practice questions by the topics actually tested at L4, with framing on what each round is looking for.
The interview loop
- 1Recruiter screen30 minutes. Background, motivation, logistics. Light technical at most.
- 2Phone screen (technical)45 minutes, one coding question. Algorithms or data structures. Expect Medium difficulty with at least one follow-up.
- 3Onsite: Coding round 145 minutes, one to two algorithmic problems. Trees, graphs, dynamic programming, two-pointer/sliding window are common.
- 4Onsite: Coding round 245 minutes, problem-solving round. Often more open-ended, sometimes systems-y (rate limiter, LRU cache, scheduler).
- 5Onsite: System design45 minutes. L4 candidates get bounded designs: URL shortener, news feed, chat, basic distributed cache. Drive the conversation, ask clarifying questions, discuss trade-offs.
- 6Onsite: Googleyness & leadership45 minutes. Structured behavioral. Conflict, ambiguity, learning from failure, why Google. Have STAR stories ready.
What Google actually evaluates
- →Code that handles edge cases without being prompted
- →Naming, structure, and clarity over cleverness
- →Thinking out loud - silent coding gets a 'no signal' rating
- →Clarifying questions before jumping to a solution
- →Honest 'I don't know' over confident BS
- →Curiosity and intellectual humility (Googleyness)
Topics tested
Algorithms
Two of three coding rounds focus here. Expect Medium difficulty with optimization follow-ups - 'can you do this in better space?'
Data Structures
Trees, graphs, hash tables, heaps. Tries appear more often than you would expect for autocomplete-style problems.
System Design
Required at L4 but bounded. Don't try to design Spanner. Pick a sensible scale, walk through the components, defend trade-offs.
Behavioral
The Googleyness round is structured. Prepare 6-8 STAR stories covering conflict, ambiguity, leadership, learning, impact.
Databases
Comes up inside system design - sharding, indexing, ACID vs BASE. Rare as a standalone topic.
Object-Oriented Design
Used in a coding round if you get a design-y problem (parking lot, elevator). Less common at L4 than at L3.
System design topics tested in this loop
Curated walkthroughs for the bounded designs that show up in Google's system design rounds. Capacity estimation, architecture, deep-dives, and trade-offs.
URL Shortener
MediumThe canonical bounded system design problem. Read-heavy, hot-key prone, and a great vehicle for hashing, caching, and capacity estimation.
News Feed
HardThe classic write-vs-read amplification trade-off. Push, pull, or hybrid fanout - and how to handle the celebrity user with 100M followers.
Chat
HardLong-lived connections, ordering guarantees, presence, and the difference between 1:1 chat and a 50K-member group.
Rate Limiter
MediumFive algorithms, three sharding strategies, one fail-open vs fail-closed decision. The bounded design that surfaces in every backend interview loop.
Video Streaming
HardEncoding ladders, adaptive bitrate, CDN economics, and the difference between live and VOD. Petabyte-scale storage meets millisecond-scale playback.
Ride-Share Dispatch
HardGeo-indexing, real-time matching, ETA prediction, and surge. The canonical geo-spatial design problem with hard real-time constraints.
Web Crawler
HardPoliteness, deduplication, freshness, and the URL frontier. The classic crawl-the-internet question that surfaces deep distributed systems judgment.
Behavioral themes tested in this loop
Sample STAR answers, common prompts, pitfalls, and follow-up strategies for the behavioral themes that decide Google's loop.
Googleyness
GoogleNot a soft round. Structured questions about collaboration, ambiguity, learning, and motivation - scored against rubrics, not vibes.
Ambiguity
GeneralTested at Google, Anthropic, OpenAI, and any senior+ loop. Strong candidates show how they get curious; weak candidates show how they get anxious.
Conflict
GeneralThe most universal behavioral question. Tested everywhere. The signal is in how you investigate the disagreement, not in how you 'won.'
Learning from Failure
MicrosoftMicrosoft's Growth Mindset core. Also tested at Google, Anthropic, and any company that screens for self-awareness. The signal is whether you actually changed.
Compensation at Google BETA
Total comp ranges, base, equity, and bonus across the levels tested in this loop. Aggregated from public sources.
Google compensation by level
5 SWE levels covered. Updated 2026-05.
Curated practice questions
333 MCQs and 229 coding challenges, grouped by topic. Free preview shows question titles - premium unlocks full content.
Algorithms · 77 MCQs
Browse all in Algorithms →Data Structures · 44 MCQs
Browse all in Data Structures →System Design · 68 MCQs
Browse all in System Design →Behavioral · 63 MCQs
Browse all in Behavioral →Databases · 49 MCQs
Browse all in Databases →Object-Oriented Design · 32 MCQs
Browse all in Object-Oriented Design →Algorithms - Coding challenges · 149 challenges
Browse all coding challenges →Data Structures - Coding challenges · 50 challenges
Browse all coding challenges →System Design - Coding challenges · 2 challenges
Browse all coding challenges →Databases - Coding challenges · 25 challenges
Browse all coding challenges →Object-Oriented Design - Coding challenges · 3 challenges
Browse all coding challenges →Practice in mock interview format
Behavioral and system design rounds reward practice with a live AI interviewer that probes follow-ups, not silent reading.
Start an AI mock interview →Frequently asked questions
How is L4 different from L3 at Google?
L3 is entry-level (new grad). L4 is mid-level with 3-5 years of experience. The bar at L4 includes system design as a required round, deeper behavioral signal, and an expectation that coding rounds will surface follow-up questions to test depth. L3 candidates often skip system design or get a much lighter version.
Does Google still ask LeetCode-style questions?
Yes, but less than the meme suggests. Coding rounds use LeetCode-style problems as a vehicle for evaluating problem solving, code clarity, and communication. Memorizing 300 problems will not help if you can't articulate your approach. Practice talking through problems out loud.
How much system design do I need to know for L4?
Enough to drive a 45-minute design of a moderately complex system (chat app, news feed, URL shortener) without freezing. You should know caching layers, load balancing, database sharding, eventual vs strong consistency, and message queues. You don't need staff-level depth on consensus protocols or custom storage engines.
What is the Googleyness interview really testing?
Three things: do you collaborate well, do you handle ambiguity, and do you actually want to work at Google. It's structured - questions are scored against rubrics, not vibes. Vague stories or 'I'm a hard worker' answers score poorly. Prepare specific incidents using STAR.
How long does the Google interview loop take?
Recruiter screen to onsite typically takes 4-8 weeks. After the onsite, hiring committee review takes 1-3 weeks, then team matching can take another 2-6 weeks. Plan for 2-4 months end to end. Use that time to keep practicing - most candidates lose momentum between rounds.
Can I retry if I get rejected?
Yes. Standard cooldown is 6-12 months depending on how close you were. Recruiters track which round was the weak signal and will sometimes coach you on what to focus on. Many engineers get hired on attempt two or three.