HackerRank launched Chakra, an AI agent that conducts technical job interviews, to general availability on Monday after six months in beta testing. The system has conducted more than 500,000 interviews during its testing phase, with companies including Snowflake, Snorkel, and Capgemini participating.
Chakra evaluates candidates beyond correct answers. The AI agent observes how candidates approach problems, asks follow-up questions based on their work, and assesses what HackerRank calls "AI fluency" - how well candidates frame problems for AI systems, judge their output, and steer toward solutions. The platform combines what previously required three separate interview rounds into a single assessment.
During a Chakra interview, candidates work through real-world code repository tasks using a canvas that includes an AI assistant. The system can ask contextual follow-up questions, such as why a candidate chose a particular approach or how their solution would change under new constraints.
HackerRank co-founder and CEO Vivek Ravisankar told TechCrunch that AI has made traditional coding assessments less useful. He said the core question for employers has shifted: "Now, because of AI, anybody can produce an artifact." Employers need to evaluate the thinking and judgment behind that artifact, not just the output.
Chakra flags suspicious activity at rates 70 percent to 80 percent lower than traditional HackerRank assessments, according to Ravisankar. He attributed this to candidates having legitimate access to AI during interviews, which reduces incentive to secretly use outside tools.
The move represents a significant pivot for HackerRank, which built its business on coding challenges. Ravisankar compared the transition to Apple moving from the iPod to the iPhone, suggesting Chakra represents the company's strategic direction forward.
The deployment raises regulatory questions. New York City requires employers using certain automated employment decision tools to conduct independent bias audits and notify candidates beforehand. Ravisankar acknowledged hiring is a regulated area and stated HackerRank has built compliance into the system. He argued that properly tuned AI can apply hiring criteria consistently across all candidates, potentially reducing bias from factors like education or background. Automated hiring tools can nonetheless inherit or amplify biases from training data, models, and criteria used to build them.
Ravisankar stated Chakra scores candidates rather than making final hiring decisions, which remain with humans.