HackerRank Makes Its Chakra AI Interviewer Generally Available After 500,000 Beta Interviews -- and Starts Scoring 'AI Fluency'

On October 5, 2026, after roughly six months in beta, HackerRank made Chakra, its AI interviewer, generally available to customers. TechCrunch reports that Chakra ran more than 500,000 interviews in beta; Runtime Wire notes HackerRank has not said whether that counts unique candidates, completed sessions or attempts. Named testers were Snowflake, Snorkel and Capgemini, plus HackerRank itself.

An enterprise example: how the workflow runs

• One session instead of three rounds. Chakra combines the recruiter screen, the take-home assessment and the engineer follow-up interview into a single AI-conducted session.
• Real repository, AI assistant allowed. Candidates work on realistic tasks in an actual code repository, with an integrated AI assistant on the canvas. Chakra watches how they work, then probes -- asking, in TechCrunch's description, "why they chose one approach over another, or how their solution would change if a new constraint were introduced."
• What is scored. Problem-solving, judgment, communication and "AI fluency": how well a candidate frames a task for the AI, evaluates what it returns, and steers it to a useful result. Hiring teams receive section-level scores, feedback and supporting evidence.
• Data touched. HackerRank's candidate notice, quoted by Runtime Wire, says the system may analyze "coding, behavioral signals, webcam images and screen activity for performance assessment or integrity monitoring."

CEO Vivek Ravisankar's case for the design: "The previous modality of evaluation was evaluating the output. Now, because of AI, anybody can produce an artifact." HackerRank says suspicious-activity flags ran 70% to 80% lower than in its traditional assessments -- unsurprising when the assistant is allowed rather than hidden.

The adoption constraint is evaluation and regulation, not technology. Ravisankar told TechCrunch that Chakra scores candidates but humans make the hiring decision, and acknowledged that hiring is regulated and that compliance is part of what HackerRank has had to build; TechCrunch notes, for instance, that New York City requires an independent bias audit and candidate notice for certain automated employment decision tools. He also said "AI is way less biased than humans, if you tune it properly" -- a company claim. Runtime Wire notes HackerRank's materials say Chakra is evaluated for fairness, but that interview volume does not establish whether its scores predict job performance or treat candidates consistently across groups. The candidate notice offers location-dependent rights, including an alternative process or human review.

Enterprise pattern (analysis): the interesting move is measuring the process, not the artifact. The same logic applies inside companies evaluating engineers who already work with coding agents. But anything that scores people from webcam and screen activity becomes a high-risk system under most emerging rules, so buyers should ask for the bias audit, the scoring rubric and the appeal path before rollout, and should check whether a "score" is in practice the decision.