Mercor is sticking with traditional whiteboarding techniques in its hiring process rather than relying on take-home tasks.

During a recent appearance on the "20VC" podcast, Osvald Nitski, the company's head of product, explained the shift away from assignments that AI could easily handle.

He noted that one interview round assesses familiarity with AI tools, while emphasizing skills in setting up experiments, statistical understanding, sound judgment, and systems design through whiteboarding exercises.

These whiteboard sessions, where applicants sketch out their reasoning, rose to prominence in the 2010s and remain common at firms including Google, Meta, and Microsoft.

Nitski highlighted concerns that AI enables excessive delegation of choices, stressing the need for candidates to demonstrate independent judgment instead of simply repeating outputs from tools like Claude.

Although Mercor incorporates AI for initial candidate screening in its talent platform, its approach to full-time roles stands apart from broader industry trends favoring AI-assisted evaluations.

Work trials and take-home projects are gaining traction as alternatives to conventional interviews, with AI-focused startups such as Lovable, Cursor, and Kilo favoring practical experiments to evaluate both technical abilities and interpersonal skills.

This strategy also sets Mercor apart from competitors, as recent reports indicate that major tech players like Google, LinkedIn under Microsoft, and Cisco are now permitting AI use during interviews, while some startups provide complete access to coding assistants.

Emily Cohen, head of people and operations at Cognition, an AI coding firm, described the rationale for adapting processes to include AI and vibe coding, comparing restrictions on AI to denying a student a calculator during a math exam and advocating for tool usage in role-relevant tasks.