Shopify's AI Coding Interview: Bring Your Own Tools

Shopify's engineering interviews include a pair programming round, and candidates report that you bring your own editor and your own AI tools to it. That makes it one of the most open AI rules of any company, and also one of the least documented. This guide separates what Shopify says itself from what candidates and prep sites report. For other companies' AI rules, see our AI coding interview overview; for the format in general, see what an AI coding round is.

AICodingRound is not affiliated with Shopify, and none of our problems are Shopify questions.

Practice the steering and checking on AICodingRound. The reports say Shopify judges how you drive the AI, test what it gives you, and explain the result. Our problems train that on real multi-file apps, with your own Claude or Codex as a coach. Try Test the API rate limiter's edges (medium, about 10 min). The beta is free and places are limited: get a place.

What Shopify says on the record

About the interviews. Shopify's careers FAQ describes its steps in general terms. For R&D applicants, it says "you'll pass this step and go directly into pair programming before heading to The Life Story." The Life Story is "a 30-60 minute conversation-style interview with a recruiter." The Craft assessment is "a 60-75 minute interview" with at least one person from your craft, plus "a brief predictive assessment." It also says: "The order and number of interviews may vary depending on the role and team you are applying for." The FAQ does not mention AI tools at all.

About AI at work. In April 2025, CEO Tobi Lütke's internal memo on AI became public. Digital Commerce 360 quotes it: "Reflexive AI usage is now a baseline expectation at Shopify," and reports that AI competency would become part of performance reviews and hiring decisions (Digital Commerce 360, April 8, 2025). TechCrunch quotes another line: "Before asking for more headcount and resources, teams must demonstrate why they cannot get what they want done using AI" (TechCrunch, April 7, 2025).

The memo is about how Shopify employees work. It is not a published interview rule, but it explains why an AI-friendly coding round would fit the company.

What candidates report

Everything in this section is reported, not confirmed by Shopify.

Hello Interview's guide is undated and says it is based on candidate reports, without naming them. The PracHub post is one candidate's first-person account, edited by the site. GreatFrontEnd's Shopify guide adds a useful caution: "Some coding sessions allow AI tools; confirm the rules for your interview before rehearsing with them" (GreatFrontEnd).

Where the reports disagree

• The kind of problem. Hello Interview describes self-contained design problems in an empty repo. The PracHub candidate describes a business system (automatic refunds) from a list of requirements. Both are builds from scratch, but they test different things: class design in the first, product reasoning in the second.
• How long. We found one candidate report of 75 minutes. Shopify's FAQ gives 60 to 75 minutes for the Craft assessment but does not say how that relates to pair programming.
• One round or two. Hello Interview reports two AI rounds. The PracHub candidate describes only the technical screen. The FAQ says the number of interviews varies by role and team.

The safest reading: expect to build something small from an empty or near-empty repo, with your own tools, for about an hour. Confirm the rest with your recruiter.

What gets judged (reported)

Hello Interview lists workflow and planning, design and extensibility, readable code, testing, how you handle bad AI output, and communication (Hello Interview). Two lines it attributes to Shopify interviewers sum it up: "We don't want the AI to work for you. We want you to tell it what to do." And: "even if the AI gives you code that works, you have to go over it and fix it."

In other words, the AI is allowed to type. You are expected to drive, to check what it typed, and to explain it.

How to prepare

Get your setup boring. Have your editor, AI tools, language runtime and test runner working before the call, and make sure you can clone a repo and push a branch to GitHub without thinking. The PracHub candidate wrote: "I wasted a lot of time setting up git at the beginning," and felt rushed for the rest of the round.

Practice starting from nothing. Create an empty repo and build a small thing in an hour: a rate limiter, a parking lot, a tail that does not read the whole file into memory. Spend the first five minutes on a short plan and a few questions before any prompt.

Make testing a reflex. Hello Interview reports that testing came up directly, including how you would design unit tests. After each chunk the AI writes, run it, add or update a test, and confirm the behavior before moving on.

Expect the requirements to change. Follow-ups like "now support many robots" or "make the capacity configurable" test whether your first design can stretch. Keep the code simple enough to change.

Narrate. Say what you are asking the AI for and why, and what you checked in its answer. A silent hour of prompting gives the interviewer nothing to score.

Practice for Shopify's round on AICodingRound

AICodingRound trains the parts the reports say get judged: extending a design that already exists, writing tests that catch real bugs, and spotting problems in AI-written code. Your own Claude or Codex works as a coach that explains but does not write the code, so you practice checking and owning every line. One note: the reported round starts from an empty repo, so also rehearse one from-scratch build with your own setup.

• Book a room every week (medium, Python, about 45 min). Add a feature with real edge cases.
• Save a cart for later (PRD) (medium, TypeScript, about 50 min). Build from a product spec.
• Stop rescheduling from regressing (medium, Python, about 40 min). Write tests that catch real bugs.
• Test the API rate limiter's edges (medium, Python, about 10 min). A short testing drill.
• Review the AI's retry PR (medium, Python, about 10 min). Practice catching what an AI got wrong.

Each one ends with a short debrief about your own work, close to the explaining you will do on the call, and a report with a level per skill (Strong, Solid or Developing), including how you used AI. Our guide to AI fluency interviews covers the steering and checking habits in more depth.

Get your reps in before the call. The beta is free and places are limited: get a place.