Key Takeaways
• Google will allow some engineering candidates to use Gemini during interviews
• The pilot program begins in the second half of 2026 in the U.S.
• Candidates will use AI during code comprehension and debugging tasks
• Interviewers will evaluate AI fluency alongside technical skills
• The move reflects broader changes in software engineering in the AI era
Google is preparing a major overhaul of its software engineering interview process by allowing candidates to use Gemini during portions of technical evaluations.
According to reports, the company will begin a pilot program in the second half of 2026 for select junior and mid level engineering roles in the United States. The initiative marks one of the clearest signs yet that major technology companies now view AI tools as a core part of modern software development rather than something to exclude from assessments.
Under the new system, candidates will be allowed to use Gemini during a “code comprehension” round. Instead of solving isolated algorithm problems from memory, applicants will analyze, debug, and optimize an existing codebase while working alongside the AI assistant.
The goal is to evaluate how effectively candidates can collaborate with AI systems in realistic engineering scenarios. Interviewers will assess skills such as prompt engineering, validating AI generated outputs, identifying errors, and refining solutions produced by the model.
Brian Ong, the company’s vice president of recruiting, said the changes are intended to make interviews more reflective of how engineering teams now operate in the AI era.
The Gemini assisted round is part of a broader redesign of Google’s hiring process. The company’s long standing “Googleyness and Leadership” evaluation will now include technical discussions focused on projects candidates have worked on previously. For more junior applicants, one traditional technical round will also be replaced with an open ended engineering challenge designed to test problem solving methodology instead of memorized answers.
Google plans to initially test the new format within its Cloud division and platforms and devices teams before potentially expanding the system globally.
Internally, the company reportedly describes the approach as “human led, AI assisted.” Rather than replacing technical ability, the AI acts as a collaborative tool that candidates must use effectively and critically.
The move comes as the role of AI in hiring continues to evolve rapidly. Surveys suggest that a growing number of job seekers are already encountering AI driven interviews and automated screening systems. However, Google’s approach stands out because it places the AI directly in the hands of the candidate instead of using it solely behind the scenes for evaluation.
This shift could fundamentally change how technical competence is measured in the software industry. As AI becomes embedded in daily engineering workflows, employers may increasingly prioritize the ability to work with intelligent systems rather than relying only on traditional coding tests.
The pilot also reflects a broader transformation happening across the tech sector. Software development is moving away from pure manual coding toward a hybrid workflow where engineers guide, validate, and refine AI generated outputs.
If successful, Google’s experiment could influence hiring practices throughout the industry and reshape expectations for the next generation of software engineers.
Your First 10 AI Skills
10 practical AI skills, copy-paste prompts and a 7-day plan to start using AI with confidence.
Download the guide →