Canva now lets candidates use AI during interviews: What this means

India Today

12-06-2025

Canva has officially changed its technical interview process by introducing AI-assisted coding, allowing and even expecting candidates to use tools like GitHub Copilot, Claude, and Cursor during interviews for engineering roles. The move is a big change in how the company evaluates software engineers, aligning the hiring process with the tools its in-house developers already use on a daily basis.Until now, Canva's coding interviews, particularly for backend, frontend, and machine learning roles, relied heavily on conventional Computer Science Fundamentals assessments. These tests, which involved solving algorithm-based problems without any external help, were no longer the day-to-day reality at Canva as engineers now regularly depend on AI tools to navigate large codebases and accelerate development.advertisementRealising that this disconnect limited the company's ability to assess true on-the-job performance, Canva has now introduced what it calls "AI-Assisted Coding" interviews. Candidates are not only allowed but also encouraged to use AI during interviews. The company says that the idea is to understand how well they can collaborate with AI tools, how effectively they break down complex problems, refine machine-generated code, and make sound technical decisions.
Talking about the trials done for this, Canva head of platforms Simon Newton said, "Our AI-assisted interviews feel more engaging for both candidates and interviewers, and they are providing strong predictive signals about candidate performance. Most importantly, they are helping us identify engineers who can leverage AI thoughtfully and effectively. Exactly the kind of people we want building the future of visual communication," he said.advertisementThis change comes amid growing debate around the use of generative AI in hiring. While some companies continue to view AI in interviews as a form of cheating, Canva says it is trying to restrict AI use is counterproductive in an industry where such tools are already an integral part of development workflows. Instead, the company is prioritising transparency and encouraging candidates to use AI openly during interviews.The company revealed that candidates are now judged on how they guide the AI, evaluate its suggestions, and maintain production-level standards in their final solution. The interviews also test how well candidates debug flawed AI code and deal with unclear product requirements."Interestingly, candidates with minimal AI experience often struggled, not because they couldn't code, but because they lacked the judgment to guide AI effectively or identify when its suggestions were suboptimal," Newton said. "We believe the future belongs to engineers who can seamlessly blend human creativity and judgment with AI capabilities," he added.