HackerRank shifts from coding tests to AI-driven candidate evaluation

The platform has already processed over 500,000 interviews, with companies like Snowflake and Capgemini participating in the trial. CEO Vivek Ravisankar describes the transition as a pivot from evaluating static answers to analyzing the cognitive process behind them. Because generative AI allows anyone to produce functional code, the company argues that traditional assessment models have lost their effectiveness. Chakra attempts to solve this by forcing candidates to explain their logic and adapt to new constraints in real-time, effectively collapsing three traditional hiring rounds into a single, comprehensive session.

Despite concerns that providing AI tools during an interview might encourage dishonesty, HackerRank reports that suspicious-activity flags dropped by 70% to 80% during testing. Ravisankar contends that by normalizing AI usage within the assessment, the incentive for candidates to secretly rely on external bots evaporates. While the system provides detailed reports and scoring, the company maintains that final hiring decisions remain with human managers. This approach, however, faces the growing challenge of regulatory scrutiny regarding algorithmic bias. As cities like New York implement mandatory audits for automated hiring tools, HackerRank claims to have built its infrastructure to meet these compliance requirements while arguing that a properly tuned AI can apply rubric criteria more consistently than a human interviewer.