DoorDash Interview Questions — Page 12

DoorDash·Machine Learning Engineer·Technical Phone Screen

May 2026

DoorDash MLE interview with a meaty system/product debugging question about order fulfillment. The kind of question that sounds like a PM problem until you realize they actually want you to think about data pipelines and service boundaries too.

A customer receives items that don't match what they ordered. Walk through every possible reason this could happen across the full pipeline, from the customer placing the order to the courier dropping it off. Cover UX issues, restaurant errors, courier behavior, system bugs, data sync problems, fraud, and edge cases like out-of-stock substitutions. Then explain how you'd prioritize investigating each category.

“This one took me a minute to even figure out where to start.”

View Post

DoorDash·Software Engineer·Technical Phone Screen

May 2026

DoorDash software engineering interview that centered on a practical coding problem tied directly to their core business. The follow-up pushed into product-adjacent territory which I wasn't fully expecting from a coding round.

Write a function that calculates a delivery driver's total daily pay from a list of delivery records, where each record might include start time, miles driven, a base rate, and a tip. Handle missing or invalid fields gracefully, apply a minimum pay floor per delivery, and return a per-day total.
Extend your solution to support a configurable 'peak hour' rule, where deliveries starting within certain time windows earn a higher pay rate. The time windows should be passed in as a parameter.

“The error handling part is where I spent most of my mental energy.”

View Post

DoorDash·Data Scientist·Onsite - Behavioral / Leadership

May 2026

Behavioral round for a senior data science role at DoorDash. The whole thing was basically one big leadership question about mentorship, with a bunch of follow-ups ready to go depending on how you answered.

As a senior data scientist, how would you mentor junior teammates? Be specific and give real examples.
Tell me about a time you mentored someone who was struggling or being defensive about feedback.
How do you mentor someone when you actually disagree with the technical direction they're taking?
When you're heads-down on your own deliverables, how do you keep mentorship from falling apart?

“This one is deceptively broad.”

View Post

DoorDash·Machine Learning Engineer·Onsite - System Design / Architecture

May 2026

System design round at DoorDash for an MLE role, focused entirely on infra for a large-scale recommendation system. No model architecture stuff, just the serving layer, which I wasn't fully expecting going in.

Walk me through the high-level architecture for serving real-time recommendations at scale, including both the online request path and any offline or batch components.
How would you design the caching layer for this recommendation system? What do you cache, what are your cache keys, how do you set TTLs, and how do you handle staleness?
How do you scale the candidate generation and ranking services to handle spiky, high-QPS traffic across multiple regions?
What are your storage choices for the feature store, embedding store, and item catalog, and what consistency guarantees do you need from each?
How do you make the system resilient to downstream failures, like the feature store or embedding service being unavailable?
What observability would you build into this system? What metrics, logs, and alerts matter most?

“I started with the online path and worked backwards to the batch side, which felt natural but I think I spent too long on the request flow before even mentioning the batch jobs.”

View Post

DoorDash·Software Engineer·Technical Phone Screen

May 2026

DoorDash software engineer round with a problem I'd never seen before. Had to code it live which added some pressure, but the problem itself was actually interesting once I wrapped my head around it.

Given three arrays representing chef skill levels, dish difficulty levels, and dish profits (all the same length), assign each chef to exactly one dish such that a chef can only cook a dish if their skill level meets the dish difficulty. Multiple chefs can be assigned to the same dish. Maximize the total profit.

“Never ran into this type before.”

View Post

DoorDash·Machine Learning Engineer·Onsite - System Design / Architecture

May 2026

DoorDash ML engineer interview focused heavily on a multi-objective ranking problem for restaurants. The whole thing was basically one long design question with a lot of follow-ups baked in, which I wasn't fully expecting.

Design a restaurant ranking model that jointly optimizes for conversion rate and expected delivery time. Walk through your candidate architectures.
How would you design the loss function for this two-objective model, and how do you weight the two heads against each other?
Once you have predictions for both CVR and delivery time, how do you combine them into a single ranking score?
How do you handle label correlation and selection bias when training this model?
What offline metrics would you track for this system, and how would you set up an online A/B test to evaluate it?

“I went straight to multi-task learning with shared layers and two output heads, which felt like the obvious move.”

View Post

DoorDash·Data Scientist·Take-home Assignment

May 2026

DoorDash data science take-home, heavy SQL focus with BigQuery/Snowflake syntax. Four tasks all in one script, all requiring CTEs, and they really wanted to see LAG and QUALIFY used correctly. Not a casual exercise.

Using CTEs and window functions, compute the city-day cold complaint rate for completed deliveries over the last 30 days. Return city, date, delivery count, cold complaint count, and complaint rate. Handle timezone conversion from UTC to city-local time and state your assumption.
Using LAG over order_events partitioned by order_id, compute ready-to-pickup and pickup-to-dropoff durations per order. Then compare median pickup-to-dropoff time between orders with a cold complaint versus without, broken out by city, for a one-month window. Use QUALIFY to keep only cities where the difference exceeds 12 minutes.
For each courier, restrict to their last 100 completed deliveries in their city up to a cutoff date. Flag couriers whose cold complaint rate exceeds twice the city median rate and who have at least 100 deliveries. Use QUALIFY both to select the last 100 deliveries per courier and to filter the outlier couriers.
Compute the P90 of pickup-to-dropoff time per city. For each courier, find the share of their deliveries that exceed their city's P90. Use QUALIFY to return the top 5 couriers per city by that share, with ties broken by higher delivery count. Explain the tie-breaking window you used.

“The timezone part is what trips people up.”

View Post

DoorDash·Software Engineer·Technical Phone Screen

May 2026

DoorDash software engineer interview focused almost entirely on one deep project walkthrough. They really mean it when they say 'go deep.' Be ready to defend every decision you made like you're presenting to a skeptical senior staff eng.

What problem were you solving in your most impactful project, and why did it matter to the business or users?
What was your specific role on the project, and what did you personally contribute versus the team?
What measurable impact did the project achieve, and how did you define and track success?
Walk me through one or two key technical decisions you made, including the trade-offs you considered around design, algorithms, data, or infrastructure.
How did you approach scaling and reliability for the project, including things like service level objectives, failure modes, and how you rolled it out safely?
Describe a risk or failure you encountered during the project and how you handled it.
If you were starting this project over today, what would you do differently and why?
If you can't recall specific details about the project, how would you go about reconstructing that context before making a decision?

“This part felt easy until I realized I was rambling about technical context without ever landing on the actual stakes.”

View Post

DoorDash·Software Engineer·Onsite - System Design / Architecture

May 2026

DoorDash system design round focused on a deceptively practical API orchestration problem. The setup was clean but the failure-handling requirements made it way more involved than it first looked.

You have three internal APIs: one that resolves a user ID to a consumer ID, one that fetches payment info given a consumer ID, and one that fetches address info given a consumer ID. Build a bootstrap endpoint that returns all four fields combined, calling the payment and address APIs in parallel after the first call resolves, with timeouts, retries with exponential backoff, and graceful partial responses when the payment or address call fails permanently. If the first call fails, the whole thing fails.

“The parallel calls part I got pretty quickly, just fan out after the first resolves and wait on both.”

View Post

DoorDash·Software Engineer·Technical Phone Screen

May 2026

DoorDash coding screen, one problem, pretty standard binary search setup but they wanted the full proof of correctness which I wasn't expecting at all.

You have an array where each element represents a pile size, and an integer H representing total hours available. Find the minimum processing rate R (items per hour) such that you can finish all piles within H hours, processing one pile at a time and rounding up fractional hours. Walk through your algorithm, prove it's correct, and give time and space complexity.

“I got the binary search angle pretty quickly.”

View Post