Aug 19: Bhairaju Pallavi is a mechanical engineer. Covid made his worst fears come true. Salary cuts, lay off uncertainty showed him the bottle necks of his field. When he tried to transition, he enrolled in an AI/ML course taught in English and dropped it within weeks, convinced the subject wasn’t the problem, he was. Then he found the same curriculum offered in Telugu. He is now in his 3rd Trimester.
His story isn’t an anecdote. It’s a pattern. Of the 14,971 students who’ve enrolled in IIT Patna Vishlesan i–Hub‘s certification programs to date, almost 60% are learning in a language other than English, up from a 0% regional-language cohort when the i–Hub first launched its AI/ML program. That number should unsettle anyone who’s spent the last three years insisting that India‘s AI gap is a skills gap. It isn’t. It’s a translation gap and it’s been hiding in plain sight because nobody bothered to ask whether “explain gradient descent” needed to happen in English to be understood.
This isn’t just a curriculum decision. It’s a bet on a version of India‘s AI future that doesn’t route through English fluency. One where a mechanic in Coimbatore, a shop owner in Nashik and a doctor in a Hindi-belt district all have equal claim to the AI economy, not a translated, second-tier version of it. Bharat, in other words, isn’t waiting for India to teach it AI in its own terms. It’s building the terms itself.
The i–Hub didn’t add regional languages as a courtesy. It rebuilt its curriculum around them. Tamil, Telugu, Marathi and Hindi versions of its AI and Machine Learning and Full Stack Development programs now sit alongside English, not as a simplified version, but as the same technical rigor, delivered in the language a learner actually thinks in.
The rest of the enrollment data reads the same way. Gaurav Kumar, a practicing doctor, enrolled in the hindi-medium program with one goal: provide affordable healthcare to his community. Sandeep Shitole, a business owner, is building AI expertise in Marathi alongside running an export business.
Neither needed English. They needed the material in a language that didn’t make them feel stupid before the lesson even started.
“Our response in both the Tamil and Telugu belts has been extraordinary. We initially expected regional language demand to be modest, a nice-to-have alongside our English
cohorts. Instead, enrollment in these languages has matched or outpaced English in some programs, which told us this wasn’t a side offering, it was the actual unmet demand. Technical fluency in English has long been treated as a prerequisite for technology education. Our enrollment data says otherwise. When we saw the enrollment split by language, it wasn’t a marginal difference, it reframed how we thought about the entire funnel” – Prof. Dr Sai Kiran Oruganti, Ceo IIT Patna Vishlesan i–Hub.
The question the i–Hub is betting on isn’t whether Indians can learn AI. The enrollment numbers already answered that. It’s whether India‘s AI economy will be built by the 10% who speak English, or by the Bharat that was always capable and was just never asked in a language it understood. At nearly 15,000 enrollments and rising, the i–Hub is betting on the latter.