The End of an Era
For decades, the whiteboard interview was the technical screen's final boss. Candidates were asked to solve complex algorithmic problems on a board, from memory, often in front of a panel of silent judges. The process was designed to test a candidate's
raw knowledge of data structures and algorithms. However, it has long been criticised for being a poor proxy for on-the-job performance. Critics argue it favours recent graduates who have just studied these topics, induces intense anxiety, and doesn't reflect how developers actually work: collaboratively, with access to documentation and tools. The disconnect between this high-pressure, artificial test and the reality of software engineering has led many to question its value.
The AI Co-Pilot Arrives
The game changed with the widespread adoption of AI coding assistants. Tools like GitHub Copilot, Gemini, and Claude can now generate, debug, and explain code with startling proficiency. With reports suggesting over 80% of developers globally now use AI to help write code, it has become an integral part of the modern development workflow. This has rendered the old style of testing obsolete. If a developer's daily job involves collaborating with an AI, why test them in an environment where such tools are forbidden? As one industry expert noted, asking a developer to code without AI today is like asking a mathematician to take a test without a calculator.
The New Interview: Human-Led, AI-Assisted
In response, companies like Google, Meta, and Canva are piloting new interview formats that embrace AI. Instead of a blank whiteboard, a candidate might be given a flawed or incomplete codebase and access to an AI assistant. The task is no longer to write perfect code from scratch, but to use the AI as a collaborator to debug, refactor, and build new features. The interviewer can often see the candidate's entire interaction with the AI—every prompt, every accepted suggestion, and every discarded idea. The focus has shifted from the final code to the process of getting there.
The Skills That Matter Now
This new format tests a different, arguably more relevant, set of skills. Companies aren't just looking for coders; they are looking for engineers who can effectively manage a powerful but sometimes unreliable tool. The key competencies now include: prompt engineering (the ability to ask the AI the right questions), output validation (spotting when the AI's plausible-looking code is subtly wrong or inefficient), and systems thinking (understanding the architectural trade-offs of the AI's suggestions). The most valuable skill is no longer coding speed but engineering judgment. The goal is to see if the candidate can lead the AI, not just follow its instructions blindly.
Implications for the Indian Tech Landscape
This global shift has significant implications for India's massive pool of tech talent. As Indian MNCs and the tech giants operating in India adapt their hiring processes, engineering graduates and professionals must evolve their skills. The surge in AI-related hiring in India highlights the demand. The era of rote memorization of algorithms for interviews is waning. Instead, aspiring engineers need to focus on practical application and developing a deep understanding of how to work with AI tools. This means building projects, practicing prompt engineering, and learning to critically evaluate AI-generated code, transforming from pure coders into AI-augmented problem solvers.
