Situation report: The coding test as we know it is on its way out. HackerRank, the developer assessment platform, is making Chakra generally available to customers on Monday. Chakra is an AI agent that runs job interviews, watches candidates while they work, and grades how they reached an answer as well as the answer itself, according to TechCrunch AI.
That’s an opportunity for employers who are tired of take-home tests anyone can finish with a chatbot. It’s also a warning sign for anyone uneasy about algorithms deciding who gets hired.
🎯 What Launched
Chakra spent about six months in beta. HackerRank says it ran more than 500,000 interviews in that time. Snowflake, Snorkel, and Capgemini all tried it, and HackerRank also used it for its own hiring.
Here’s how an interview works, per TechCrunch AI:
- Real work, not puzzles. The candidate gets a task inside a real-world code repository.
- AI is allowed. They work in a canvas with an AI assistant built in.
- Live follow-ups. Chakra watches what the candidate is doing and asks questions in context, like why they picked one approach over another, or how the solution changes if a new constraint shows up.
- Report to the hiring manager. When the interview ends, the hiring manager gets a detailed report.
🧠 What It Measures
The big change here is what gets scored. HackerRank wants to capture critical thinking, judgment, and something it calls “AI fluency.” That means how well a candidate frames a problem for AI, judges what it gives back, and steers it toward a solution.
“The previous modality of evaluation was evaluating the output,” co-founder and CEO Vivek Ravisankar told TechCrunch. “Now, because of AI, anybody can produce an artifact.”
He’s right about that. If an AI can write the code, the code doesn’t tell you much. How a person thinks while using the AI is the part that’s still hard to fake.
⚡ Tactical Impact
Ravisankar says Chakra replaces three hiring steps with one interview:
- Recruiter screen
- Take-home assessment
- Follow-up interview with an engineer
For hiring teams, that could cut weeks from the process. For engineers who currently lose hours to early-round interviews, it’s time back.
Cheating is the other surprise. You’d expect an open AI assistant to make cheating easier. HackerRank says it did the opposite. Suspicious-activity flags came in 70% to 80% lower than in comparable traditional HackerRank assessments, though the number varied by geography and seniority. Ravisankar’s explanation is simple: if AI is already allowed, there’s less reason to sneak in an outside tool.
🔄 A Bet Against Its Own Core Business
This part is the most interesting to me. HackerRank launched at TechCrunch Disrupt in 2012 and built its business on coding challenges. It’s backed by Y Combinator and now has more than 3,000 business customers, including Amazon, Nvidia, Clay, and Replit, plus a community of over 30 million developers.
Chakra cuts against the product that got it there. Ravisankar compared the shift to Apple moving from the iPod to the iPhone. The old product still has value, but the new one shows where things are going. “Chakra is going to be the headline,” he said.
It’s rare for a company to say out loud that its flagship product is losing relevance. It’s a smart move, though. If HackerRank doesn’t replace its own coding tests, someone else will.
⚠️ Risks
- Bias claims need proof. Ravisankar says “AI is way less biased than humans, if you tune it properly,” since it applies the same rubric to every candidate. But a consistent rubric doesn’t make a system unbiased. Automated hiring tools can inherit or amplify bias from their data, models, and criteria.
- Regulators are watching. New York City already requires employers using certain automated hiring tools to get an independent bias audit and to tell candidates beforehand. Ravisankar says HackerRank has built for these requirements.
- Humans keep the final call. Chakra scores candidates but doesn’t decide who gets hired. The idea is that AI handles the structured evaluation while human interviewers judge fit and answer questions about the team and role.
🔭 Outlook
Expect competitors to follow fast. Once employers see interviews where AI use is allowed and scored, banning AI in technical assessments will look outdated. The open question is whether hiring managers will trust an AI-written report on a candidate’s judgment, and whether regulators will accept it. Full details are in TechCrunch AI’s original report.