HackerRank is releasing its AI interviewer, called Chakra, to customers on Monday following roughly six months in beta. The AI agent conducts interviews, watches candidates work, and evaluates both final answers and the problem-solving process.
During testing, Chakra completed over 500,000 interviews with companies like Snowflake, Snorkel, and Capgemini, alongside internal testing by HackerRank. Co-founder and CEO Vivek Ravisankar stated that the tool shifts evaluation from static outputs to assessing critical thinking, judgment, and AI fluency.
Chakra combines multiple hiring steps
A Chakra interview resembles real-world work rather than a standard coding test. Candidates work in a canvas with a real-world code repository and an integrated AI assistant. Chakra asks follow-up questions about approach choices and new constraints based on active work context.
The process condenses a traditional three-step sequence into a single interview. Ravisankar noted that suspicious-activity flags dropped by 70% to 80% in Chakra assessments compared to traditional HackerRank formats, though rates varied by geography and seniority.
Automated tools face regulatory oversight
HackerRank built its business on coding challenges, but Ravisankar views Chakra as the future direction for the company. The Y Combinator-backed startup serves more than 3,000 business customers and has a community exceeding 30 million developers.
Chakra scores candidates rather than making final hiring choices, leaving ultimate decisions to humans. Ravisankar argued that properly tuned AI applies criteria more consistently than humans, though automated hiring tools face scrutiny and bias audits from regulators such as New York City.
