For years, giving candidates coding tests were hiring's closest thing to an objective filter. Give candidates the same algorithmic problem/coding tests, score it, move the top scorers forward, hire 1 of them. Clean, scalable, defensible.
Then AI arrived and made the whole thing irrelevant overnight.
Today, any candidate with access to ChatGPT, Copilot, or Cursor can generate fake resumes in 2mins, or clear a HackerRank/Leetcode-style tests. Tech leaders know this. Most are quietly scrambling for an alternative.
Very few have found one that actually works at scale. Utkrusht AI has a solution to the current tech hiring problem.
The Vadodara-based tech startup, founders themselves engineers, which recently closed a seed funding round, has built what is the world's first “Watch-them-Work” assessment platform – a fundamentally different approach to technical hiring that does not fight AI, but instead uses it as a signal.
Instead of testing candidates on traditional coding tests which doesn’t mirror the actual conditions of the job, Utkrusht AI takes a different approach to put all candidates inside a live production environment and gives them real on-the-job tasks, and then allows you to watch HOW a candidate actually works.
For eg:
A fullstack candidate debugs a broken API and deploys a fix
An SRE candidate writes an incident response runbook
An AI engineer improves embeddings in a production chatbot
etc.
These tasks are carefully designed from engineers from Oracle, Microsoft, Google, etc., takes 30-45 minutes, the session is fully recorded, candidates can use any tools they want, AI included, so all of their thinking and decisions are captured.
What the platform captures is not just the output. It captures everything in between:
how the candidate broke down the problem,
what decisions they made and abandoned,
where and how they used AI,
whether it helped or hurt,
how they handled ambiguity when there was no clean answer in the documentation
This is precisely what no other candidate assessment solution captures today.
Coding tests show a score. Take-home assignments show a finished product. Watch-them-Work shows the thinking behind it – which is increasingly the only thing that separates strong engineering hires from expensive mistakes.
The numbers back it up
Utkrusht has assessed over 11,000 candidates on from customers across India, Europe, and USA. Companies using the platform report a ~70% reduction in time-to-hire, and usually hire in less than 14 days.
Plus, ~90% candidates complete the assessment during working hours – not on weekends, not reluctantly, because a 30-min real task is a better use of time than a four-hour take-home.
Your dashboard shows a ranked shortlist of the top 5-10 candidates worth interviewing, with detailed reports scored across technical execution, judgment, thought process, communication, and AI usage, plus a recorded session for every candidate and a clear hire/no-hire recommendation.
Vedang Manerikar, Founder of Unravel Tech, said his team was spending 15 hours a week on resume screening and first-round calls before switching.
Tanay Shah, Founder of Pardy Panda Studios who hires developers a lot, eliminated code pairing sessions entirely - his senior developers were spending 20+ hours a week on them.
Why this matters now
The companies that figure out how to hire for judgment rather than syntax speed in the next 12-18 months will build meaningfully better tech teams than those that don't.
Utkrusht is currently offering a free trial - no credit card required, setup under five minutes. For teams not actively hiring, sample candidate reports are available on request, showing exactly what a strong engg/developer hire looks like in the AI era.
Visit utkrusht.ai to check out..
Disclaimer: This is a sponsored article. All possible measures have been taken to ensure accuracy, reliability, timeliness and authenticity of the information; however Outlookindia.com does not take any liability for the same. Using of any information provided in the article is solely at the viewers’ discretion.
