Not too long ago, job interviews involved at least one HR representative pretending to care. There would be some awkward small talk, a polite “how was your weekend?”, maybe even a forced laugh about how you’re hiding out in your current employer’s cafeteria. Now, companies are starting to skip all of that and go straight to the point. In some cases, straight past people altogether, often with bizarre consequences. Which is how you end up in situations where correct answers are apparently incorrect, and the interviewer may or may not know who they work for.
AI interviewers are quietly becoming part of the hiring process may be great news for efficiency—especially for roles where one has to go through thousands of applications—but it’s terrible news for anyone who relied on being a personality hire.
Some companies across Silicon Valley have started outsourcing the first round to tools like TestGorilla, while others are going a step further. Listen Labs, for instance, runs automated AI “interviews” at scale, clocking more than a million conversations for clients including Microsoft and Perplexity AI. Which means your carefully crafted, AI-assisted answers may now be evaluated by…another AI.
Of course, like all things tech, it works perfectly. Except when it glitches. Or freezes. Or cuts you off mid-sentence. Or simply moves on, leaving you desperately discussing your leadership skills to a screen that has long stopped listening. There are also different versions of this experience. While platforms like CodeSignal will have you talking to what is essentially a very serious looking AI headshot, others like Humanly have gone all in on fully anthropomorphic avatars that look like they have too many opinions.
And then there is the slightly uncomfortable part. These systems are trained on existing data, which means they come with biases. Apologies in advance if your name has more than three syllables, if your accent requires basic listening skills, or, God forbid, if you are a woman who sounds a little too confident before 11 am. Not that you would ever know. You will simply be “not a fit”.
Globally, companies are testing all of this, and it is only a matter of time before it becomes standard in the Indian job market. Why have HR scan applications when a bot can do it faster, without breaks, and without needing to Google “ice breakers for fresh candidates” five minutes before the call? So, in the spirit of adapting to this brave new workflow, we have put together a practical guide to cracking the AI interview round. Or at least, surviving it long enough to maybe speak to a human again.
Rehearse your answer. Then rehearse repeating it. Then rehearse repeating the repetition.
Because nothing says excellent communication skills like saying the same sentence three times while a bot buffers like it’s emotionally overwhelmed.
Eliminate all background noise
A passing scooter, a barking dog, or your mother asking whether you want chai will instantly re-categorise you as “unsuitable for corporate environments”.
Pause strategically, but not for too long
A brief pause is good. A three-second silence, on the other hand, may be interpreted as “candidate has left the chat”.
Speak slowly as if you’re leaving a voicemail for your grandmother
You say “I led the project”, it hears “I left the project”, and now you have professional commitment issues. Perhaps permanently.
Maybe…reconsider your name
Studies show that AI systems rate candidates with ethnic-sounding names lower before you’ve uttered a single word. So, if your name looks like a spelling mistake on Google Docs, you’re already a few steps behind.
Your biggest weakness should be “being a bit of a perfectionist”
Everything else is now classified information, and the AI is logging your weaknesses somewhere in a server in Texas. Take your flaws to your grave.
Aim for an algorithmically neutral smile
Think friendly LinkedIn profile picture. Any emotion beyond that may be considered unstable.
If the bot freezes, keep talking
Maintain eye contact with the buffering screen as you talk about your five-year plan. It’s excellent practice for the AI manager you will probably be reporting to a few years from now.
Do not make an ‘AI is taking over our jobs’ joke
It won’t laugh. It will remember. And honestly, this is not the moment to test its sense of irony.
Remember that if you’re a woman, the margin for error is…tighter
Too assertive? Aggressive. Too soft? Lacks leadership. The bot may not understand gender, but the data set it learned from definitely did not either. Sigh.
Be prepared to be ruthlessly cut off mid-sentence
The bot has no emotional attachment to your college fest story, your internship trauma, or your personality. It will interrupt you with the efficiency of someone who has never had to be polite
Finally, remember: you are being evaluated for “culture fit”
...by a machine that has never experienced culture, work, or the dread of a sudden Microsoft Teams notification popping up with “Can we hop on a quick call?”.
