
Thousands of travellers and trekkers go to Leh and Kargil every year. In the government schools scattered across both, far fewer people ever think to visit.
In episode 4 of Stories from the Schools of Bharat, Rohit Prakash, co-founder of iDream Education, sat down with Yogesh, from iDream Education’s operations team, who went there for the first time this year, to train teachers and check what the schools’ classrooms needed. He landed in Leh, into a climate far colder than he was used to, acclimatised for a day, and set out the next morning to survey a school’s infrastructure.
The children were the first thing that struck him, ‘a different level altogether,’ he says, and the teachers were just as respectful. That’s where the story actually starts: not the terrain, but what the team decided to build once they saw what these classrooms needed. Internet connectivity turned out to be the single biggest obstacle, especially in Kargil, so personalised adaptive learning had to work whether or not a connection ever reached the school.
Here’s what Yogesh told Rohit.
First visit to Ladakh, and why PAL got built there

The interaction with the children, Yogesh says, was something else entirely.
“They were [Ladakhi] children of a different level altogether. Very respectful, and the teachers were very respectful too.”
– Yogesh, Operations Team, iDream Education
That’s where the implementation began. The team set up personalised adaptive learning (PAL) there to give Leh Ladakh’s school children a more personalised kind of education.
No internet, not even for the monthly data sync

Ask Yogesh what the single biggest challenge was, and the answer comes fast: internet connectivity.
Most of iDream Education’s smart classroom and PAL lab setups run almost entirely offline.
The content and software work without needing a live connection, and teachers and students use it that way day to day.
But the team still needs to check the data now and then – how much a lab is being used, how teachers are running it, how far each child’s learning has moved.
So there’s usually a simple rule built in. Once a month, sometimes once a week, occasionally once every fortnight, teachers connect their devices to the internet long enough for everything to sync. That rule assumes there’s internet to connect to.
In Leh, some areas have it. In Kargil, where this implementation sits, there’s none nearby at all. Getting a usable signal means coming some distance down from the mountain.
A Raspberry Pi and a router, doing the internet’s job
The fix Yogesh’s team built runs on a Raspberry Pi, a small, low-cost computer, paired with a router. Neither needs the internet to work. The router creates a local network, and the Raspberry Pi sits on that network, talking to every tablet in the room.
Each school gets around 25 to 30 tablets, and the learning content itself lives on an SD card in every tablet. A child can pick up any tablet in the room and start learning straight away.
Here’s how the data actually moves once a child finishes on a tablet:
- The tablet connects to the router’s local WiFi, and whatever the child has done syncs to the Raspberry Pi.
- The teacher’s phone connects to that same local network, and the teacher’s app pulls all of that data in.
- Once the teacher’s phone reaches an actual internet signal, everything moves across to iDream Education’s central system.
This solves a problem most fully offline PAL labs run into.
If a child’s data only ever lives on the one tablet they used yesterday, picking up a different tablet today usually means starting the same chapter over again. Because Kargil’s tablets are constantly talking to the same Raspberry Pi, a child’s progress follows them onto whichever tablet they pick up next.
The same setup handles something else: keeping content current as the National Education Policy (NEP) reshapes syllabuses. Updated content goes onto the Raspberry Pi’s SD card, and it syncs out to every tablet in the room automatically, with nobody needing to update each device by hand.
When a school is too remote for that content to arrive directly, the team sends it to the headmaster by email, through Google Cloud. The headmaster downloads it the next time they’re near a signal, saves it onto the Raspberry Pi’s card, and the same automatic sync takes care of the rest.
Badges, a final test, and chocolate for the teachers who finished
Training ran over three days, with teachers arriving from different schools across the district. Yogesh says they were curious rather than wary about the new setup – most of them wanted to know what it was.
One task stuck with him. Teachers were asked to complete the final test inside the app for any one chapter in their subject, the same test their students would eventually take. Plenty of them got competitive about it.
“Whichever teacher earned a badge in their subject’s chapter, we gave them chocolate too, so there’d be some extra excitement.”
– Yogesh, Operations Team, iDream Education
The Raspberry Pi, the SD cards, the training, none of it changed what personalised learning is meant to do for a child. What it removed was the excuse that altitude and missing internet get in the way first.
iDream Education partners with schools in terrain like this to keep personalised learning running on local infrastructure, built to work with or without a signal. If your CSR programme or NGO needs an implementation for a similarly disconnected region, get in touch to talk through what that would take.




