In the AI Era, Shopify Is Investing in Junior Engineers—Not Cutting Them - CoderPad
Shopify grew its internship program tenfold to recruit developers who’ve grown up with AI and to gauge their coding and judgment in real projects. It evaluates candidates across three modes—no AI, AI optional, AI mandatory—to test hand-coding skills, tool discernment, and AI-driven development.
- Shopify grew its internship program from ~100 to over 1,000 spots, betting on junior talent instead of cutting it in the AI era
- Candidates are tested in three modes—no AI, optional AI, mandatory AI—to assess raw coding skill, judgment, and AI-driven building
- AI has collapsed grunt work like unit tests, refactoring, and learning new frameworks into near-instant tasks, freeing engineers to prototype more boldly
- Even Shopify admits there's no proven formula yet for which skill mix (manual coding vs. AI fluency) actually predicts job success
Shopify ramped its internship program from about 100 spots a year to over 1,000, and plans to keep 1,000 interns on board this cycle. Farhan Thawar, Shopify’s VP of Engineering, says early-career hires bring fresh perspectives—especially in an era where AI tools are woven into every stage of development. Interns aren’t siloed; they spread across teams, boosting curiosity and energy. The four-month stint doubles as a “two-way interview,” letting Shopify assess coding chops, AI fluency and cultural fit, while interns get real projects, daily in-office interaction and a shot at full-time offers.
Shopify treats AI as part of every engineer’s toolkit rather than a specialized discipline. Every pull request still bears human ownership. The hiring process now tests candidates under three conditions: no AI (can you code from scratch?), optional AI (do you know when to lean on tools?) and mandatory AI (can you build complex features under tight deadlines by prompting and validating output?). That framework comes from the University of Waterloo, and Thawar admits it’s imperfect—no one yet knows which mix of skills best predicts on-the-job success in this new world.
Thawar highlights three ways AI changed day-to-day work: managers can drop into code in spare minutes via tools like Claude Code or Cursor; tasks such as writing unit tests or refactoring legacy modules no longer require multi-day deep dives; and exploring new languages or frameworks happens instantly, so ambitious engineers can prototype without fear. His advice to students and new grads remains the same: build tools that solve real problems, put them on GitHub and invite feedback. With AI handling much of the grunt work, there are fewer excuses not to create something useful right now.