Uber Software Engineer mock interview

4-5 onsite rounds including a unique machine coding round (build a small runnable system) and Uber-specific system design (ride dispatch, surge pricing, multi-region serving).

  • 4-6 weeks
    Process timeline
  • Machine coding
    Unique round
  • Dispatch / surge
    SD flavor
  • Free
    First mock

Why Uber's machine coding round is the dimension most candidates underestimate

Per prepfully.com's Uber SWE 2026 guide, algo.monster's Uber guide, and interviewing.io's Uber question collection, Uber's machine coding round is a unique dimension that most candidates underestimate. You're given 60-90 min to build a small runnable system — parking lot reservation, in-memory KV store with TTL, library management — with clean class hierarchy and code that compiles on first try.

This isn't whiteboard pseudocode. It's not algorithmic optimization. It's practical engineering judgment compressed into 60-90 min: design class hierarchy, implement core operations, ensure thread safety where relevant, write working code that runs. Candidates who optimize for cleverness over working state fail this round. Practice the format specifically — build incrementally, compile/test frequently, prioritize working over complete.

System design is also Uber-specific. Common questions reference Uber's actual product: ride dispatch matching, surge pricing pipelines, multi-region rate limiting. Generic 'design Twitter'-style answers without Uber framing lose senior signal. Reference Uber's actual scale (millions of rides/day, multi-region presence) when discussing trade-offs — that's the dimension graded.

What the loop looks like

Round-by-round

  1. 01

    Recruiter screen

    Pre-onsite
    30 min

    Experience, motivation, role fit, level expectation. Uber explicitly calibrates by years of experience here.

  2. 02

    Online assessment OR live coding screen

    Pre-onsite
    60-90 min

    1-2 algorithm problems. Codility-style automated assessment for some pipelines, live phone screen for others.

  3. 03

    Technical phone screen

    Pre-onsite
    60 min

    Medium-to-hard coding problem with complexity discussion. Production-quality code expected on first pass.

  4. 04

    DSA coding round

    Onsite
    45-60 min

    LeetCode-style algorithmic problems. Edge cases and complexity narration graded explicitly.

  5. 05

    Machine coding (UNIQUE to Uber)

    Onsite
    60-90 min

    Build a small runnable system with clean code. Low-level design — class hierarchy, interactions, runnable on first compile. Tests practical engineering judgment over algorithmic cleverness.

  6. 06

    System design

    Onsite
    60 min

    Uber-specific problems: ride dispatch matching, surge pricing pipelines, multi-region availability with eventual consistency. Scale numbers (millions of rides/day) expected.

  7. 07

    Behavioral

    Onsite
    45 min

    Ownership, teamwork, decision-making. Stories about driving outcomes through ambiguity. Level-calibrated — senior+ stories must show cross-team scope.

Question bank

Real interview questions reported by candidates

  • System design

    "Design Uber's ride dispatch system. Cover driver-rider matching algorithm, geo-indexing, ETA calculation, and what happens when a driver cancels mid-trip."

    Source · interviewing.io Uber questions, hellointerview.com Uber guides, designgurus.io Uber SD
  • System design

    "Design surge pricing. Real-time supply/demand analysis, region-level pricing decisions, and propagation latency to driver/rider apps."

    Source · interviewing.io Uber questions, prepfully.com Uber SWE 2026
  • System design

    "Design a multi-region rate-limiting service for the Uber API. Cover region failover, eventual consistency on rate counts, and abuse detection."

    Source · hellointerview.com Uber guides 2026
  • Coding

    "Machine coding: build a parking lot reservation system. Class hierarchy, concurrency-safe reservations, runnable on first compile. 60-90 min to a working solution."

    Source · prepfully.com Uber SWE 2026, algo.monster Uber guide
  • Coding

    "Machine coding: implement a in-memory key-value store with TTL eviction. Thread-safe operations, runnable tests included."

    Source · interviewquery.com Uber SWE 2026, algo.monster Uber guide
  • Coding

    "DSA: graph problem with non-trivial constraints (shortest path with budget, or visit-all-with-K-backtracks). Expected to discuss complexity at multiple optimization levels."

    Source · prepfully.com Uber SWE 2026, Glassdoor Uber SWE reports
  • Behavioral

    "Tell me about a time you owned a complex system in production. What did you learn from running it?"

    Source · techprep.app Uber 2026, prepfully.com Uber SWE 2026
  • Behavioral

    "Describe a time you made a difficult engineering decision under ambiguity. How did you decide?"

    Source · techprep.app Uber 2026
  • Behavioral

    "Tell me about a time you collaborated across teams to solve a problem. What was your role?"

    Source · interviewquery.com Uber SWE 2026
Common signals to fix

What gets you rejected at this level

  • Machine coding that doesn't compile / run

    #1 Uber-specific failure mode. Machine coding round expects code that runs on first compile within the 60-90 min window. Pseudocode answers, syntax errors you don't catch, or solutions that 'would work but I ran out of time' fail. Practice the format: build incrementally, compile/run frequently, ensure working state before adding features.

  • Generic system design without Uber framing

    Uber's SD round expects Uber-flavored framing — dispatch matching, surge pricing, real-time geo-indexing. Generic 'design Twitter'-style answers without Uber context flag missing company-specific senior judgment. Reference Uber's actual scale (millions of rides/day, multi-region presence) when discussing trade-offs.

  • Behavioral stories below level scope

    Uber explicitly calibrates by level. New grad stories about class projects work at L4-equivalent. Senior+ stories must show cross-team scope — driving decisions across multiple teams, mentoring senior engineers, raising engineering practice across the org. Stories below this scope flag down-level.

  • Skipping complexity narration

    Same as Google / Amazon — DSA rounds at Uber expect you to call out time/space without prompting. Multiple rounds of forgetting flags missing senior signal.

  • OA underperformance

    Codility-style OA in some pipelines is auto-graded. Failing it ends the loop. Practice the format specifically — bare-editor environment, time pressure, edge cases graded.

How Applr's AI mock interview tracks Uber's rubric

Applr's Uber mock includes the machine coding round format — runnable system in 60-90 min, code-quality + working-state both graded. The system design round prompts Uber-flavored framing (dispatch, surge, multi-region) and grades scale-number coherence.

Behavioral rounds are level-calibrated; the recruiter-equivalent intake at setup confirms target level (new grad / mid / senior / staff) and the mock loads the appropriate scope expectations. Stories below the target scope get flagged with rewrites.

Adjacent loops

Related practice

FAQ

What's the machine coding round really like?

Per prepfully.com's Uber SWE 2026 guide and algo.monster's Uber guide: machine coding is a low-level design round where you build a small runnable system in 60-90 min. Examples: parking lot reservation system, in-memory key-value store with TTL, library management system. The bar is code that compiles and runs on first try, with clean class hierarchy and reasonable interaction model. Practice the format: build incrementally, compile/test frequently, prioritize working state over completeness.

How Uber-specific is the system design round?

Per interviewing.io's Uber question collection and hellointerview.com's Uber guides: Uber's SD round is heavily Uber-flavored. Common questions: design ride dispatch (driver-rider matching, geo-indexing, ETA), design surge pricing (real-time supply/demand, region-level decisions), design multi-region rate limiting. Generic 'design Twitter'-style answers without Uber context lose senior signal. Practice these specific Uber-style designs.

How does Uber's leveling work in interviews?

Per interviewquery.com's Uber 2026 guide: Uber explicitly calibrates expectations by level during the screen. New Grad / Entry: 2 coding rounds + light design. Mid-level (2-5y): coding + full system design. Senior (5-8y): platform-scale system design with multi-region, incident narratives, mentoring impact. Staff+: cross-org design, cost/performance trade-offs, long-term technical strategy. Recruiter sets level expectation in the screen — be honest about your YoE.

Does Uber sponsor visas?

Yes for engineering roles in major US offices (SF, NYC, Seattle). International offices in Amsterdam, Bangalore, Singapore offer alternative pathways. Confirm with recruiter early — visa coordination can extend the process by 2-4 weeks for complex situations.

How does Uber compare to Airbnb / Coinbase?

All three are tier-1 unicorns with strong engineering interviews. Uber: machine coding round + Uber-specific SD (dispatch, surge), behavioral on ownership. Airbnb: code review round + vague-by-design SD + Host interviews on Core Values. Coinbase: 90-min CodeSignal filter + 2-onsite (behavioral + SD), no crypto knowledge required. Pick by which company / domain interests you most.

What's the salary band at Uber for SWE?

Per levels.fyi and Glassdoor 2025 data: Mid-level SWE roughly $250-320K total comp; senior $350-500K. Equity component significant (Uber is public). Comparable to other tier-1 US unicorn bands at equivalent levels.

Ready to practice for Uber?

Start free →