Canva Software Engineer candidates report a recruiter screen, AI-assisted coding, system design, code review and a values-based behavioral round. Rendering and image-processing design prompts come up repeatedly.
$103K
Avg. Base Comp
$130K
Avg. Total Comp
3-7 rounds
Typical Rounds
Not reported
Process Length
Canva's Software Engineer interview, as candidates describe it, is built around practical product engineering rather than puzzle-solving alone. Several recent candidates report an AI-assisted coding round where you build something concrete with tools like Cursor or Copilot. One was asked for a text-rendering CLI that outputs a PNG, and another was given a deliberately vague text-and-font-to-image prompt. Interviewers appear to watch how you clarify requirements and make architectural decisions on the fly. One candidate was told that a nearly finished solution was not enough, so practice thinking aloud and finishing a working result.
System design leans toward Canva-flavored problems. Candidates mention a rendering system where users download their output and a scalable image upload system with processing and storage optimizations. They also mention a Google Docs-like collaborative editor that has to handle conflicting edits. Another candidate reports an e-commerce site. Expect to defend tradeoffs, and prepare for follow-ups on CDNs, caching and reliability under load.
Later rounds often add a language-fluency or unassisted coding session, a code review or release-plan discussion, and a behavioral round tied to Canva's values. One candidate reports Java fundamentals such as equals/hashCode, synchronization and checked versus unchecked exceptions, plus writing HashMap functions from scratch. Behavioral prompts covered disagreeing with a product manager or designer, delivering uncomfortable news and missed team goals. Some candidates report a long final loop of back-to-back sessions, so plan your energy. Outcomes were mixed: there were some offers, one at a lower level than targeted, and several rejections.
Synthesized from 29 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Canva process.
I went through six interviews over several months. The recruiter reached out on LinkedIn with impressive promises about a smooth process, and the initial call with them was genuinely positive. I then interviewed with a hiring manager, which went well, followed by an AI interview where I was asked to code a solution for an NP-hard problem. That's where things got tricky—I used a cheaper AI model to help solve it, and only at the very end of all rounds did they tell me that choice had hurt my scoring. That felt unfair since no one mentioned model cost or quality during the interview.
After that came a relatively short forty-five-minute system design interview, then a leadership round focused on code review and release planning, and finally a Java coding challenge without AI assistance where I had to build a comments system. By that point I'd invested significant time, and I genuinely thought I'd had solid conversations throughout. But then came the ghosting. For weeks, nothing. When they finally responded, they said they were "not sure" and offered vague feedback without a clear rejection. The silence made me realize they were probably running all candidates' interviews in parallel, making final decisions based on comparative performance rather than individual feedback. The whole process felt designed to extract as much insight as possible before deciding—interview everyone to the end, then rank and select. I left disappointed and exhausted, with no clear understanding of why I didn't advance.
Prep tip from this candidate
If using AI for the AI-assisted coding round, use a capable model (GPT-4 or better) and be transparent about your tool choices—they evaluate your ability to collaborate with AI effectively, not just get the answer. Expect a long multi-stage process and don't assume silence means a decision; be prepared for ambiguous feedback or extended ghosting even after final rounds.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Canva
Write a query that returns all neighborhoods that have 0 users.
| Question | |
|---|---|
| Address Schema | |
| Hurdles In Data Projects | |
| Search Ranking | |
| Duplicate Rows | |
| Generate Shopping List from Recipes | |
| International e-Commerce Warehouse | |
| Unified Event Pipeline | |
| Electricity Supply | |
| k-Means from Scratch | |
| Statistically Significant Test | |
| 2nd Highest Salary | |
| Merge Sorted Lists | |
| Top Three Salaries | |
| Closest SAT Scores | |
| Subscription Overlap | |
| Monthly Customer Report | |
| Random SQL Sample | |
| Top 3 Users | |
| Find the First Non-Repeating Character in a String | |
| String Shift | |
| Rolling Bank Transactions | |
| Customer Orders | |
| Minimum Change | |
| Raining in Seattle | |
| Comments Histogram | |
| Download Facts | |
| P-value to a Layman | |
| Flight Records | |
| Bagging vs Boosting |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report a 30-minute conversation about background and motivation for joining Canva. Have a clear reason for wanting Canva and a concise story about your experience.
Many candidates describe a coding round where AI tools such as Cursor or Copilot are allowed. Reported tasks include a text-to-image rendering tool with a vague spec and a text rendering CLI that outputs a PNG. Interviewers appear to value clarifying questions and tradeoff decisions. One candidate says a complete working result mattered.
Candidates report prompts such as a rendering system with downloadable output, image upload with caching, a collaborative editor and an e-commerce site. Reported sessions ran about 45 minutes. Be ready to discuss CDNs, caching and scalability, and to justify your choices.
Later technical sessions reportedly include implementing methods or HashMap-style functions without AI help, reviewing someone else's code, and discussing a release plan or delivery risk. Candidates describe these as closer to day-to-day work, so practice explaining issues and tradeoffs clearly.
Several candidates report a behavioral or values round, sometimes straight after technical sessions in one sitting. Questions covered disagreements with a product manager or designer, delivering difficult news and missed goals. Outcomes varied from offers, in one case at a lower level, to rejections.
