Got the offerSSE L5 at Uber
- Difficulty
- Process took
- 3-4 Weeks
- Rounds
- 5
- Format
- Remote
- Applied via
- Company Website
How it went
- Do not overstretch your prep timeline, revision is key.
- Hard work + consistent momentum beats last-minute cramming.
What they would tell you
- Do not overstretch your prep timeline, revision is key.
- Hard work + consistent momentum beats last-minute cramming.
How to prepare
DSA:
-
Practice a wide range of graph, DP, and array problems. System Design:
-
Break down into FRs, NFRs, API design, data pipelines.
-
Use tools like Kafka, Flink, Redis, and SQL databases.
Their background
Years of Experience- 7 years 8 months Current Company- MAANG Bangalore location Current Compensation- 75 LPA
Round by round
- 1
Coding Round 160 min
In my first round, I was asked to find the minimum cost to travel in a grid where moving in the arrow’s direction is free and any other move costs 1. I started with brute-force DFS, discussed its high complexity with the interviewer, briefly considered DP, and then moved to a graph-based approach. I modeled the grid as a weighted graph with 0/1 edges and used Dijkstra’s algorithm to solve it.
- 2
Coding round 260 min
- 3
Machine Coding Round60 min
I had to implement expiry counter, basically a class with methods to add elements and fetch counts, but with each element expiring after t seconds. I first went with a priority queue + hashmap idea to manage expirations, but the interviewer didn’t seem too satisfied. Then I suggested a simpler binary search approach, just keep a mapping of each element to its list of timestamps and use bisect to quickly find valid ones.
- 4
System Design Round60 min
In my system design round (taken by a Senior Engineer), I was asked to design a real-time surge pricing system that shows surge multipliers for city regions. The problem had two parts, a near real-time use case (updates every 1–2 minutes for the last 20 mins) and an analytics use case (hourly aggregates after 24 hours). I proposed consuming ride/driver events via Kafka, using geohashing to bucket regions, and processing streams with Flink. For real-time results, I suggested writing aggregates to Redis for fast lookups, while for analytics, hourly rollups would be stored in Postgres. We also discussed scalability, fault tolerance, and fallback strategies.
- 5
Managerial Round60 min
- Started with questions about my current and past company projects.
- Asked me to use a virtual whiteboard to explain a impactful project I had led.
- Wanted details of all challenges (technical + non-technical).
- He then described the role, product vision, and upcoming AI use cases.
- Asked why I wanted to leave my current role and what I was looking for.
What came up
A candidate-reported account, lightly edited. Interview processes change by team and date.