Google India Interview Questions 2026: Coding, System Design, and Googliness

Google India SDE interview process (L3/L4/L5):

1. Resume Screen: Google's ATS and recruiters look for: CGPA above 8.0 for freshers from top-tier colleges, competitive programming achievements (ICPC Asia regionals, Google Code Jam, Kickstart, Codeforces 1800+), relevant projects at meaningful scale, previous product company or research lab experience.

1. Phone Screen (1-2 rounds): 45-60 minutes each; 2 LeetCode medium-hard problems per round; conducted via Google Meet + a shared Google Doc (no IDE); the interviewer expects you to write clean, bug-free code in the document and narrate your approach throughout.

3. Onsite Loop (4-5 rounds, all virtual since COVID): - Rounds 1-2: Coding (2 LeetCode medium-hard problems per round; graphs, trees, DP, backtracking, bit manipulation) - Round 3: System Design (L4 and above: 'Design YouTube', 'Design a distributed key-value store', 'Design Google Docs collaborative editing') - Round 4: Googleyness and Leadership (collaboration, ambiguity, 'what would you do if you disagreed with your manager?') - Round 5 (L5 and above): additional coding or systems design

1. Hiring Committee (HC): all interviewers submit written feedback; the HC reads the full packet and makes a decision independent of individual interviewers' recommendations.

Most common Google India coding patterns:

Arrays and strings:

• Longest substring without repeating characters (sliding window: maintain a window with a HashSet; expand right, shrink left when duplicate found)
• Trapping rain water (two pointer: for each position, the water above it = min(maxleft, maxright) - height[i])
• Merge intervals (sort by start, then merge overlapping by updating end = max(end, current.end))
• Find all anagram groups (group by sorted key: 'eat' and 'tea' both map to 'aet')
• 3Sum (sort, then for each element use two pointers for the remaining two)

Trees:

• Serialize and deserialize a binary tree (BFS level-order encoding)
• Binary tree right side view (level-order BFS, take last node of each level)
• LCA of two nodes (recurse: if current node is null or equals either target, return current; if both subtrees return non-null, current is the LCA)

Graphs:

• Number of islands (BFS or DFS on 2D grid, mark visited by setting to '0')
• Course schedule II (topological sort with Kahn's algorithm: track in-degree of each node)
• Word ladder (BFS: each word with one character changed is an edge)

DP:

• Edit distance (2D DP: min of insert, delete, replace)
• Regular expression matching (2D DP with '.' and '*' handling)
• Coin change II (count the number of combinations that sum to target)

Google India system design questions (L4 and above):

1. 'Design YouTube' (the classic L4/L5 system design question at Google): Video upload pipeline: user uploads raw video to GCS (Google Cloud Storage); a Pub/Sub event triggers a transcoding pipeline that converts to multiple bitrates and resolutions (360p, 720p, 1080p, 4K); thumbnails auto-generated; video metadata stored in Spanner; transcoding completed event updates video status to 'published'. Video serving: CDN (Google's own CDN infrastructure) caches video segments close to users; adaptive bitrate streaming (HLS or DASH): the player selects the bitrate based on available bandwidth; video is chunked into segments (2-10 seconds each) for partial fetching. Recommendation: two stages: candidate generation (retrieve top 1000 videos using collaborative filtering or a two-tower neural network), ranking (score with a more complex engagement-prediction model).

1. 'Design a distributed key-value store (like BigTable/Bigtable)': consistent hashing for key distribution across nodes, replication factor of 3 (one leader, two followers per shard), LSM-tree storage engine (write to MemTable, flush to SSTable when full, compact periodically), MVCC (Multi-Version Concurrency Control) for snapshot reads.

1. 'Design Google Docs collaborative editing': CRDT (Conflict-free Replicated Data Types) for merge-free concurrent edits, operational transformation as an alternative, WebSocket for real-time sync, periodic snapshots + event log for recovery.

Googleyness and Leadership round:

What 'Googleyness' means: Google defines it as: intellectual humility (you are comfortable being wrong and learning from it), genuine collaboration (you share credit, you do not hoard information), comfort with ambiguity (you can work effectively without a clear spec), and an outsized impact mindset (you think about how to help others, not just yourself).

Common Googleyness questions:

1. 'Tell me about a time you disagreed with your manager and what you did.' Google wants to see: you raised the disagreement respectfully, with data or reasoning (not emotion), and you committed to the decision once it was made even if you were not fully convinced. They do not want: you ignored the manager, or you never disagreed.

2. 'Tell me about a time you had to work with someone who had a different working style than yours.' Show intellectual humility: you adapted to them rather than requiring them to adapt to you. Show curiosity: you tried to understand why they work the way they do.

3. 'Describe a situation where you had incomplete information and had to make a decision quickly.' Show structured uncertainty management: you identified what information would most change your decision, made a reasonable estimate of the missing data, made the decision, and built in a review mechanism.

4. 'What is something you have changed your mind about significantly?' Growth mindset. The best answers are about a technical conviction or a professional belief, not a preference.

Google Hiring Committee (HC):

The HC is a group of senior Googlers who review the interview packet without having been in the interview. They evaluate:

1. Coding ability: clean, correct code; edge cases handled; time/space complexity correct
2. Problem solving: did the candidate reach the optimal solution? Independently? With good communication?
3. System design quality (L4+): sound design, scale reasoning, trade-off analysis
4. Googleyness: genuine collaboration, comfort with ambiguity, intellectual humility, growth mindset

HC votes: SH (Strong Hire), H (Hire), LH (Lean Hire), LN (Lean No Hire), N (No Hire), SN (Strong No Hire). Most Google offers require consensus around H or above with no SN.

Google India total compensation (2024-25, Hyderabad and Bangalore): - L3 (SDE-1): 35-45 LPA base + 20-35 LPA RSUs annualised + signing bonus = 60-90 LPA TC - L4 (SDE-2): 45-65 LPA base + 35-60 LPA RSUs annualised = 90-130 LPA TC - L5 (Senior SDE): 65-90 LPA base + 60-100 LPA RSUs annualised = 130-200 LPA TC Google's RSU vesting: 25% per year (much more predictable than Amazon's back-loaded schedule).