Google India | L4 | Rejected

google logo
google
SDE IIPune/Hyderabad6 years
May 31, 20254 reads

Summary

I detail my 5-month-long Google L4 interview process for Pune/Hyderabad, covering multiple coding rounds and a Googleyness round. Despite successfully solving several algorithmic problems and receiving positive feedback on some rounds, I was ultimately rejected.

Full Experience

Hi Folks,

I wanted to share my Google interview experience for the L4 position (Pune/Hyderabad). I have around 6 years of experience, and the entire process spanned nearly 5 months, with all rounds conducted via Google Meet. While I don’t recall all questions precisely, I’ll share what I can.

Early Jan:

Recruiter from XWF reached out. I requested 10 days for preparation time as I was already practicing on LeetCode.

Mid Feb – Online Screening:

Given a graph and a minimum distance, determine if a path exists from source to target where edges exist only if the distance is less than or equal to threshold.
Follow-up: Instead of a given threshold, find the minimum threshold that allows reaching the target. Example A-->B (2), B-->C(3), A-->C(5), if source and target is A and C, answer would be 3.

Solved part 1, fell short on part 2 (likely solvable via Dijkstra).

End of March – Onsite Rounds:

Round 1:

Given match results, determine possible player rankings.
Example 1: A beats B, B beats C and A beats C. Ranking of A is 1, B is 2 and C is 3.
Example 2: A beats B, B beats C and C beats A. We cannot determine any one's ranking.
Solved with modified topological sort.

Feedback: Recruiter gave me exteremely postive feedback. My assumption on rating: Strong Hire.

Round 2:

We have 27 cards. In this we have 3 sets of 9 cards each. Each card will have a number from 1 to 9 and a colour red, green or blue. We have 4 such sets. So essentially we have 4*27 = 108 cards. A person is given 12 cards from this deck of 108 cards. We have to find if it is a winning combination. Winning combination is possible when we can split the 12 cards into 4 groups of 3. A group is valid when they have all 3 cards with same number and colour or if it forms a sequence of numbers but also of same colour. Was able to solve it using brute force but may have made some syntax errors.

Feedback: Recruiter feedback was neutral. My assumption on rating: Lean Hire.

Round 3:

Started with basic colouring permutations. We are given n nodes and m colours, we need to find out how many ways we can colour these n nodes using m colours. This question had two follow up in which the nodes were connected with some undirected edges but had no cycles.

Follow-up 1: Assuming nodes are connected with some edges, no adjacent nodes can have the same color. Find out the number of ways to colour the graph.

Follow-up 2: Find number of ways to colour the graph if no two nodes at a distance of 2 can share the same color.

Solved first two, missed edge cases in the third.
Feedback: Recuriter said feedback was negative. My assumption on rating: Lean No Hire. She said based on these rating, I may be needed to give an additional coding round but for now she will schedule my googleyness round.

Mid April – Googleyness Round:

Scenario-based discussion around managing unrealistic deadlines as a tech lead. Was able to answer most of the questions with clarity.

Feedback: Final feedback was delayed after the googleyness round even after multiple follow ups. Recruiter reached out in april end and said that she has no feedback yet for the googleyness round. There was silence again so I mailed google candidate support. Last week got feedback from recruiter that googleyness round had a positive feedback with minor mis steps. My assumption on rating would be hire.

Overall, she said that the case was too weak but she will update me in case an additional round is needed. No official rejection yet, but it's safe to assume that I’ve been rejected.

Takeaways:

Interviewers were kind and respectful. The 5-month-long process and lack of closure were frustrating but guess Google does not really care about this as they have too many talented people applying daily.

FYI:

I purposely scheduled all my rounds in night timings and all my interviewers were non-Indian. With all due respect to fellow Indians, from what I heard, Indian interviewers generally have unrealistic expectations so took a gamble with that and I felt it did pay off to some extent.

Interview Questions (7)

Q1
Graph Path with Distance Threshold
Data Structures & Algorithms

Given a graph and a minimum distance, determine if a path exists from source to target where edges exist only if the distance is less than or equal to threshold.
Follow-up: Instead of a given threshold, find the minimum threshold that allows reaching the target. Example A-->B (2), B-->C(3), A-->C(5), if source and target is A and C, answer would be 3.

Q2
Player Ranking from Match Results
Data Structures & Algorithms

Given match results, determine possible player rankings.
Example 1: A beats B, B beats C and A beats C. Ranking of A is 1, B is 2 and C is 3.
Example 2: A beats B, B beats C and C beats A. We cannot determine any one's ranking.

Q3
Winning Combination of Cards
Data Structures & Algorithms

We have 27 cards. In this we have 3 sets of 9 cards each. Each card will have a number from 1 to 9 and a colour red, green or blue. We have 4 such sets. So essentially we have 4*27 = 108 cards. A person is given 12 cards from this deck of 108 cards. We have to find if it is a winning combination. Winning combination is possible when we can split the 12 cards into 4 groups of 3. A group is valid when they have all 3 cards with same number and colour or if it forms a sequence of numbers but also of same colour.

Q4
Graph Coloring Permutations
Data Structures & Algorithms

Started with basic colouring permutations. We are given n nodes and m colours, we need to find out how many ways we can colour these n nodes using m colours. This question had two follow up in which the nodes were connected with some undirected edges but had no cycles.

Q5
Graph Coloring with Adjacent Constraint
Data Structures & Algorithms

Assuming nodes are connected with some edges, no adjacent nodes can have the same color. Find out the number of ways to colour the graph. (This was a follow-up to 'Graph Coloring Permutations' where nodes were connected with some undirected edges but had no cycles.)

Q6
Graph Coloring with Distance-2 Constraint
Data Structures & Algorithms

Find number of ways to colour the graph if no two nodes at a distance of 2 can share the same color. (This was a follow-up to 'Graph Coloring Permutations' where nodes were connected with some undirected edges but had no cycles.)

Q7
Managing Unrealistic Deadlines
Behavioral

Scenario-based discussion around managing unrealistic deadlines as a tech lead.

Preparation Tips

I requested 10 days for preparation time as I was already practicing on LeetCode.

I purposely scheduled all my rounds in night timings and all my interviewers were non-Indian. With all due respect to fellow Indians, from what I heard, Indian interviewers generally have unrealistic expectations so took a gamble with that and I felt it did pay off to some extent.

Discussion (0)

Share your thoughts and ask questions

Join the Discussion

Sign in with Google to share your thoughts and ask questions

No comments yet

Be the first to share your thoughts and start the discussion!