HackerRank launches an AI interviewer to score how engineers work with AI

Why it matters

HackerRank is moving its hiring pitch from coding-test results to observed problem-solving and AI use. The approach could make assessments closer to engineering work, while putting more weight on how employers validate and use AI-generated scores.

HackerRank made its AI interviewer Chakra generally available on October 5th, asking employers to assess how candidates work through a software problem and use AI. The launch tests co-founder and CEO Vivek Ravisankar's argument for skills-based hiring: when AI can produce working code, interviews may need to assess how candidates solve problems and use AI. TechCrunch reported that HackerRank says Chakra conducted more than 500,000 interviews during roughly six months of beta testing. Snowflake, Snorkel and Capgemini were among the testers.

Ravisankar came to hiring technology after seeing how conventional interviews could screen out capable engineers. HackerRank's company history says Ravisankar and co-founder Hari Karunanidhi had served on interview panels and watched candidates get filtered by resumes and credentials while people they considered capable were overlooked. Ravisankar had another reason to question the process: a campus interview went badly for him, a personal experience that helped shape the founders' early effort to improve interview preparation. Their first business, InterviewStreet, offered paid mock interviews; they later moved toward coding assessments for employers.

HackerRank built its business around coding challenges. Now HackerRank argues that the familiar measure of whether someone produces the right output has lost force as AI makes producing an artifact easier. "The previous modality of evaluation was evaluating the output," Ravisankar told TechCrunch. "Now, because of AI, anybody can produce an artifact."

From coding test to observed work

In Chakra, a candidate tackles a task based on a real-world code repository in a shared coding environment that can include an AI assistant. The system can ask why the candidate chose a particular approach, how they assessed AI-generated output, or what they would change if a constraint shifted. HackerRank says it evaluates problem-solving, judgment, communication and what it calls AI fluency: the ability to frame a task for AI, assess its response and guide it toward a useful result.

If AI tools are part of daily engineering work, excluding them from the interview can test an increasingly artificial version of the job. Letting candidates use AI shifts the task toward evaluating their decisions and review of the output. HackerRank's premise is that observing this work gives employers evidence a polished answer cannot.

HackerRank says the product can combine a recruiter screen, a take-home assignment and an engineer follow-up into one interview. That is Ravisankar's description of the intended process, not a published, independently measured comparison of hiring time or cost. HackerRank's product materials describe a report for hiring teams with section-level scores, feedback and supporting evidence; its candidate guide says Chakra does not make the hiring decision, which remains with the employer.

HackerRank's reported beta volume leaves open what counts as an interview: the company has not, in the sources reviewed here, specified whether the number means unique candidates, completed sessions or attempts. Interview volume also does not establish whether Chakra's scores predict job performance or treat candidates consistently across groups. Those are different tests from whether the software can conduct a session at scale.

The judgment call belongs to employers

In hiring, a score can shape a person's access to a job even if a human formally makes the final decision. HackerRank's candidate notice says its AI features may analyze coding, behavioral signals, webcam images and screen activity for performance assessment or integrity monitoring. It also describes rights that may apply depending on a candidate's location, including requesting an alternative process or accommodation and, in some cases, human review or an explanation of AI use.

HackerRank's public materials say Chakra is evaluated for fairness and alignment with human experts. Those are company claims; the launch materials do not establish independent validation of Chakra's scoring against hiring outcomes. Employers adopting it will still need to decide how much weight to give its rubric, what evidence they expect hiring managers to review, and how candidates can challenge a result they believe is wrong.

For Ravisankar, the shift is a return to the problem that started HackerRank: finding a better way to identify engineering ability than relying on conventional interviews and credentials. HackerRank now wants its software to conduct and evaluate part of the interview itself. HackerRank says AI can help employers see more of a candidate's reasoning. Whether that produces better hiring depends on the quality of the evidence Chakra captures and the judgment employers apply to its scores.