A company known for using coding challenges to assess and hire developers is now letting its customers try out an AI interviewer. Chakra is the name of that agent, and it handles interviews, watches candidates work, and judges both their answers and how they arrived at them. After roughly six months in beta, the startup is opening up Chakra to all users on Monday, and during testing it has already held more than 500,000 interviews.
The Numbers Behind Chakra
More than 500,000 interviews have already taken place while Chakra was being put through its paces, with Snowflake, Snorkel, and Capgemini among the firms that gave it a trial run. HackerRank also tested the product inside its own organization.
The startup says suspicious-activity flags were 70% to 80% lower in Chakra interviews than in comparable traditional HackerRank assessments, though the rate varied depending on factors such as geography and seniority.
What Chakra Actually Does
Chakra’s interview format resembles actual coding work rather than a standard test. Candidates receive a real-world code repository task, and they complete it inside a canvas that features an AI assistant. Throughout the session, Chakra draws on the context of the candidate’s actions to pose follow-up questions. These queries might cover why a particular approach was selected, or how the solution would adapt under a fresh constraint.
HackerRank refers to this skill as “AI fluency”, which involves framing a problem for AI, judging its output, and steering it toward a solution. The interview aims to catch these kinds of signals, including critical thinking and judgment, that are harder to measure otherwise.
“The previous modality of evaluation was evaluating the output,” HackerRank co-founder and CEO Vivek Ravisankar said in an interview. “Now, because of AI, anybody can produce an artifact.” The question for employers, he said, becomes whether they can understand the thinking and judgment that went into producing it.
Why Companies Are Using It
For a while now, businesses have relied on AI to handle job interviews. Voice agents and similar automated systems take on the task of screening candidates, speeding up the hiring process. At the same time, people applying for jobs have picked up their own AI tools to aid them during interviews, often without employers finding out about it.
The hiring process has been transformed by Chakra, which merges what was once three distinct rounds into one interview. An AI interviewer takes care of the technical assessment, leaving human interviewers free to decide whether they actually want to work with a candidate and to address questions about the company, team, and role.
TechCrunch was told by Ravisankar that Chakra exists to rank candidates, not settle on who gets hired, with humans keeping that duty. He made the comparison inside the company to Apple’s move from the iPod to the iPhone, a shift where the earlier product retains worth while the newer one shows where the startup sees the market going.
The Cheating Question
It would appear that letting candidates use AI during an interview would make cheating simpler. Instead, the firm discovered that suspicious-activity flags fell 70% to 80% lower in Chakra interviews when compared with similar traditional HackerRank assessments.
TechCrunch was told by Ravisankar that letting candidates use AI cuts down on the drive to sneak in outside tools that might supply answers during an interview.
The Bias Argument
Ravisankar says AI can manage the more structured portions of an interview by following rules given by the employer. He contends that an AI system can be told to apply the same scoring guide to every applicant without being affected by things like their background or education.
“AI is way less biased than humans, if you tune it properly,” Ravisankar said.
The case made for consistent application of criteria is not entirely persuasive. Doing so does not automatically remove bias from an AI system. Tools used for hiring can carry over or magnify biases found in the data, models, and criteria behind them, which is why regulators keep a close watch on how they are employed in employment decisions.
The Regulatory Pushback
Employers who use automated hiring tools are coming under regulatory review. One example is New York City, which mandates that firms deploying specific automated employment decision tools must first have them audited independently for bias and must also notify applicants before they are put into service.
Ravisankar admitted that hiring falls under regulated territory and said following those rules is part of what HackerRank has had to construct for.
What Comes Next
Chakra is a wager placed against the sort of technical evaluation work that HackerRank built up over many years. Its old offering mostly checked whether developers could solve coding problems accurately. Ravisankar holds that AI has rendered that approach less effective for measuring engineering skill.
“Chakra is going to be the headline,” he said. “It’s going to be the way that we’re going to move forward.”
TechCrunch Disrupt is where HackerRank first appeared, marking its launch in 2012. Since then, the Y Combinator-backed company has grown to serve more than 3,000 business customers, among whom Amazon, Nvidia, Clay, and Replit are included, while also building a developer community that exceeds 30 million people worldwide.
The firm is wagering that AI can transform both how interviews take place and what employers can gauge. The issue is whether that assessment holds up to scrutiny for fairness.
Key Facts Box
- Chakra: HackerRank’s AI interviewer, generally available Monday
- Beta duration: Around six months
- Interviews conducted in testing: More than 500,000
- Companies that tried it: Snowflake, Snorkel, Capgemini
- Suspicious-activity flags reduction: 70% to 80% lower in Chakra interviews
- Customers: More than 3,000, including Amazon, Nvidia, Clay, and Replit
- Developer community: Over 30 million worldwide
A large-scale test is underway, showing how much AI has advanced in recruiting. Hundreds of thousands of interviews have already taken place during beta, and the company is now putting its method to the test on a wide scale. The issue of whether AI can accurately gauge thinking and judgment is being put to the proof.
Source material: “HackerRank’s AI interviewer offers a glimpse into what job interviews could become,” TechCrunch.
Get the Notebook.
The day's best stories and every fresh verdict, in plain English, in your inbox by seven. One email a day, no more.