HackerRank announced Chakra, an AI interviewing agent that observes candidates as they solve real coding tasks, asks context-aware follow-ups, and scores not only final answers but also the candidate’s process and judgment. After about six months in beta, the company made Chakra generally available to customers on Monday.
How Chakra works
A Chakra interview places candidates in a workspace connected to a real-world code repository and an integrated AI assistant. As candidates work, Chakra uses the surrounding context to ask clarifying questions — for example, why a particular approach was chosen or how a solution would change under a new constraint. The system aims to measure signals that are harder to capture with traditional tests, including critical thinking, judgment, and what HackerRank calls “AI fluency”: how well a candidate frames problems for AI, judges its output, and guides it toward a solution.
Results from beta testing
HackerRank says Chakra conducted more than 500,000 interviews during its beta. Test users included companies such as Snowflake, Snorkel, and Capgemini, and the product was also piloted internally.
According to HackerRank, suspicious-activity flags were 70–80% lower in Chakra interviews than in comparable traditional HackerRank assessments, though the rate varied by geography and seniority. The company argues that providing candidates with AI during the interview reduces the incentive to covertly use external tools for answers.
Impact on the hiring process
Vivek Ravisankar, co-founder and CEO of HackerRank, told TechCrunch that Chakra can change the structure of hiring itself. Tasks that previously required multiple rounds — recruiter screening, take-home assessment, and engineer follow-up — can be consolidated into a single Chakra interview that both evaluates and interacts with the candidate.
Ravisankar emphasized that Chakra scores candidates but does not make final hiring decisions; human interviewers remain responsible for the ultimate choice. The company positions AI as handling structured, repeatable evaluation criteria consistently so human interviewers can focus on fit, collaboration, and answering candidate questions about team and role.
Business context
Founded at TechCrunch Disrupt in 2012 with support from Y Combinator, HackerRank built its business on coding challenges and technical assessments. The startup now serves more than 3,000 business customers, including Amazon, Nvidia, Clay, and Replit, and claims a developer community of over 30 million.
Chakra represents a strategic shift away from HackerRank’s earlier output-focused assessment model toward evaluating the process and judgment behind engineering work.
Bias, oversight and regulation
Ravisankar argued that a properly tuned AI can be less biased than humans because it can apply the same rubric to every candidate. However, automated hiring tools can inherit or amplify biases present in their training data, models, or evaluation criteria, and that has drawn regulatory attention.
For example, New York City requires certain automated employment decision tools to undergo independent bias audits and mandates candidate notice prior to use. Ravisankar acknowledged that hiring is a regulated area and said compliance with such requirements is part of what HackerRank built into Chakra.
Conclusion
Chakra extends AI’s role in hiring from assistance to active evaluation, offering detailed reports and structured scoring based on how candidates work as well as what they produce. While HackerRank highlights potential benefits — richer assessment signals and fewer cheating flags during pilots — the broader adoption of such systems raises ongoing questions about algorithmic bias, transparency, and regulatory compliance.
