RepostedDoorDash logo

DoorDash · Phone screen → Onsite

Rejected

DoorDash Senior SWE Onsite Interview Experience: 1 Experience

Software engineer · Full-timeMaster's degreeReported Nov 16, 2026
Shared byjanebeingjane

A 2026 DoorDash senior engineer onsite over two days: photo-feed design, a hiring manager round, a new AI coding task and a known debugging round.

At a glance

  • Process: coding phone screen → two-day onsite: project deep dive with system design, hiring manager, AI-assisted coding, debugging (report 01)
  • Asked recently: an Instagram-style feed design, a restaurant onboarding AI coding task, a round-robin dasher debugging task (report 01)
  • Difficulty: the candidate rated it average; the new AI coding task took most of the time
  • Practise first: Social News Feed With Live Engagement, Fix a Traffic Router and Add Consistent Hashing

Coverage: One experienced engineer with a Master's degree, changing jobs, interviewing for a senior engineer role, who came through a recruiter. The report covers a phone screen and a four-round onsite over two days. It gives July to September 2026 as the interview period (the onsite was in late September) and went up on November 16, 2026.

What came up

The onsite had a project deep dive with system design, a hiring manager round, an AI-assisted coding round on a new task, and a debugging round on a well-known exercise. The candidate thinks the AI coding round sank the loop.

Topic Reports Last seen
System design: a photo-sharing social feed, after a project deep dive 1 of 1 Nov 2026 (report date)
Hiring manager behavioral round 1 of 1 Nov 2026 (report date)
AI-assisted coding: a restaurant onboarding class with five calls 1 of 1 Nov 2026 (report date)
Debugging: a round-robin dasher picker 1 of 1 Nov 2026 (report date)

Questions we found

System Design

Question Reports Last seen
Design a photo-sharing app: estimate traffic, draw the data model, fan-out on read versus write, very popular accounts, images stored apart from their metadata (practice) 1 Nov 2026 (report date)

Project Coding

Question Reports Last seen
Fix the bugs in a round-robin dasher picker and run the sample (practice) 1 Nov 2026 (report date)
Implement a restaurant onboarding class: five internal and external calls must all succeed, and the restaurant ID and onboarding ID are different values (the report doesn't list the five calls, so there is no practice version yet) 1 Nov 2026 (report date)

Behavioral (BQ)

Question Reports Last seen
A time you disagreed with others on your team (practice) 1 Nov 2026 (report date)
A project of yours that failed (practice) 1 Nov 2026 (report date)

1 detailed candidate timeline, loops, and screens

01: Senior engineer, two-day onsite, not moved forward

Interview period: 2026 Q3 · Report date: November 16, 2026 · Role: Senior software engineer · Degree listed: Master's · Channel: Recruiter · Candidate's rating: Neutral, average difficulty · Result: Not moved forward

Phone screen

A coding question the candidate no longer remembers, other than that it was a commonly reported one. They passed.

Onsite, split over two days

  1. Project deep dive and system design. About 20 minutes on a past project, then 40 minutes of design. The prompt was a photo-sharing app like Instagram, which the candidate hadn't seen in earlier reports. They estimated traffic, drew the data model, and went deep on fan-out on read versus fan-out on write, how to handle accounts with huge followings, and keeping image files separate from their metadata. The candidate felt this went fine. FastPrep's practice version is Social News Feed With Live Engagement.
  2. Hiring manager. Behavioral questions, including a time you disagreed with others and a time a project failed. The candidate's advice: talk about your highest-impact projects and show the scope expected of a senior engineer. They felt this round didn't go well.
  3. AI-assisted coding. The candidate expected a known question but got a new one: a restaurant onboarding class with five functions, some internal and some external. Onboarding counts as successful only if all five calls succeed, and the code has to record both a restaurant ID and a separate onboarding ID. They used their own AI tool while sharing their screen. Understanding the task took a long time, and because the HackerRank page wouldn't let them copy the task text out, they retyped the five functions by hand to give the tool context. The tool's answer was long and they ran out of time before running the tests. Their lesson: work out the approach yourself and explain it to the interviewer first instead of handing everything to the AI.
  4. Debugging. A commonly reported exercise: a round-robin picker for dashers with several bugs. Find and fix them, then run the sample. The consistent hashing extension wasn't asked. FastPrep's practice version is Fix a Traffic Router and Add Consistent Hashing.

Result

The rejection came the next day.

Keep practicing

Sources: Anywhere useful interview info can be found.

0 total reactions. Your reactions: none.
0 comments

Comments

0

No comments yet — be the first to share what you know.