Uber SDE II Interview Experience (Bangalore): A Second Attempt That Turned Into an Offer
Updated July 17, 2026
The candidate, with about 2.6 years of experience and holding an SDE-1 title at an e-commerce company, had previously gone through an Uber onsite loop about 1.3 years earlier, in December 2022, and had not cleared the system design round, despite solid ratings elsewhere in that loop. The same recruiter later reached back out about an SDE-2 opening, at a point when the candidate was already looking to leave a difficult work environment and was separately progressing through a Google interview process.
This second Uber process opened with an online coding assessment, added an unplanned technical phone screen, and then moved into a four-round virtual onsite loop spanning coding, machine coding, system design, and a hiring-manager conversation. The candidate described the process as well organized and quick, with regular recruiter updates, though an internal delay in matching to a team pushed the final offer out by roughly two weeks after the loop concluded.
How the process went
Recruiter re-engagement
The recruiter who had run the candidate's earlier, unsuccessful Uber loop reached out again about a separate SDE-2 opening.
Online assessment
A timed coding test on Codility with four problems (one easy, two medium, one hard) and a stated cutoff around 60%.
Technical phone screen
An additional screening round, not originally part of the process, focused on a single hard grid-based DSA problem; the recruiter noted the feedback was mixed but allowed the candidate to continue to onsites.
Virtual onsite loop
Four back-to-back virtual rounds: a DSA coding round, a machine coding/low-level design round, a system design round, and a hiring-manager/behavioral round.
Team matching and offer
The original role's headcount had closed by the time the loop finished, so the candidate was matched to a different team, adding about ten days before the offer was extended two days later.
Online Assessment
General DSA problem-solving under time pressure
- Four Codility problems of mixed difficulty (one easy, two medium, one hard), scored against a roughly 60% cutoff.
Technical Phone Screen
Grid-based algorithmic problem solving
- A single hard, grid-based DSA problem where the candidate proposed a binary-search-based approach; the interviewer's expected solution was a different, more direct approach.
Virtual Onsite 1 — Coding
Data structures and algorithm design
- A 'median of a data stream' style problem.
- A harder follow-up redefining the median as any value within the range bounded by the two nearest powers of two around the true median, requiring a constant-time, constant-space solution; the candidate solved it using counters bucketed by power-of-two ranges.
Virtual Onsite 2 — Machine Coding
Low-level design and concurrency
- 'Design Facebook' as an open-ended low-level design prompt, expecting object-oriented design patterns, working implementations of three core features, and handling of concurrency via multithreading.
The prompt was given with minimal framing, leaving scope definition largely up to the candidate.
Virtual Onsite 3 — System Design
High-level system design and trade-offs
- Design a stock price indicator/tracking system, including discussion of functional and non-functional requirements, database choice and schema, and scalability approaches.
Virtual Onsite 4 — Hiring Manager / Behavioral
Past experience and leadership-principle-style behavioral questions
- Standard behavioral questions about the candidate's prior work and experience, framed around Uber's leadership principles.
Key takeaways
- An unsuccessful onsite loop is not always a permanent rejection — the same recruiter reconnected roughly a year later about a different opening.
- For grid-based or optimization-heavy DSA problems, be ready to justify time/space trade-offs explicitly, since interviewers may be anchored to a specific expected complexity class.
- Treat machine coding and system design as distinct skill sets to prepare separately: one leaned on object-oriented design patterns and concurrency, the other on schema and scalability trade-offs.
- Positive signal from every round does not guarantee a fast offer — internal headcount or team-matching constraints can add real delay even after a strong loop.
- Running more than one company's process in parallel can provide leverage and a fallback if one process stalls or a role closes.
Practice a Uber interview
Rehearse out loud against the kinds of questions in this story — with an AI interviewer that asks follow-ups.
Practice this interviewSource
The questions and process facts come from the candidate's public write-up, linked below. The retelling above is our own summary.
Candidate's public write-up on LeetCode Discuss