How the Platform Works
A plain-English account of how this platform is built, where your data lives, and why a certificate from us can be trusted. Written for students, parents, schools — and any engineer who wants to look closely.
Static by Design
The whole site is delivered as pre-built pages. That is a deliberate engineering choice, not a limitation.
Every page you load is a plain, pre-rendered HTML document served directly from a content delivery network. There is no application server interpreting requests, and no database sitting behind the website waiting to be queried. That single decision removes an entire category of failure and attack: there is no login surface on the public site to brute-force, no query layer to inject into, and nothing that falls over under sudden load. When a competition announcement drives a spike in traffic, static files simply keep serving.
It also makes the platform fast in the places that matter most. A student on a modest phone on a patchy connection in a small town gets the same document, from the same edge cache, as someone on fibre in a metro. There is no server round-trip to render a page, so the first paint depends on the network rather than on how busy our infrastructure happens to be. Accessibility of opportunity is a design goal, and page weight is part of that.
The interactive parts of the platform — the Study Hub, practice tests, paper generation, certificate verification — are implemented as JavaScript that runs in your browser, not as services that run on ours. Everything is dependency-free vanilla JavaScript: no framework runtime to download, no build step between the source and what you receive. The code that runs on your device is the code in the repository.
The Study Hub Runs on Your Device
The most private place to store a student’s learning history is the student’s own device. So that is where we keep it.
When you use the Study Hub, your progress is written to your browser’s local storage under four keys — ith_stats, ith_bookmarks, ith_goal and ith_last. Your practice scores, streaks, goals, weak topics and bookmarks live in those keys, on your device. They are never transmitted to us, because there is no endpoint to transmit them to. We could not read your study history if we wanted to.
The practical consequences are worth stating plainly. You do not need an account to use the Study Hub, so there is no password to leak. Clearing your browser data clears your progress, and your progress does not follow you to another device — that is the honest trade-off of local-only storage, and we think it is the right one for a tool used by children. If you share a device, be aware the next person sees the same local progress.
The same principle governs the rest of the platform. The site sets no cookies of its own, runs no analytics, embeds no advertising or social tracking pixels, and loads no third-party JavaScript whatsoever. We do not build behavioural profiles of students, because we do not collect the data that would make it possible.
What Actually Leaves Your Browser
Claiming “no tracking” means nothing without saying exactly which outside services are involved. There are two.
Google Fonts
Our typefaces (Cormorant Garamond, EB Garamond, Inter and Noto Serif Devanagari) are served from Google Fonts, so your browser requests those font files from Google. No account or identifier is sent by us.
FormSubmit
When you submit a contact or notification form, the form posts to FormSubmit, which relays it to our inbox as email. Only what you type into that form is sent, and only when you press submit.
Nothing Else
That is the complete list. No analytics, no advertising networks, no session recording, no social widgets, no chat SDKs, no A/B testing services, no CDNs loading executable code.
Verifiable
You do not have to take our word for it. Open your browser’s developer tools, load any page, and inspect the network tab. Every request should be to this domain or to Google Fonts.
We would rather publish a short, checkable list than a vague privacy promise. If you find a request on this site that is not on this list, please report it to us — we would treat it as a defect.
Why a Certificate From Us Can Be Checked
A certificate is only meaningful if a stranger can confirm it without asking us.
Every certificate we issue carries a unique Certificate ID in the form ITH-YEAR-COMPETITION-CODE, together with a QR code printed on the certificate itself. The QR code does not contain the result; it encodes the address of our public verification page with that certificate’s ID attached. Scanning it with any ordinary phone camera opens the verification page and checks the ID automatically.
Anyone can perform that check — a parent, a class teacher, a school admissions office, a scholarship committee, a university years from now. They do not need an account with us, our permission, or any special software. The verification page reports the participant, the competition, the category and the issue date exactly as recorded, or clearly states that no certificate matches the ID. A forged certificate fails the check, because an ID we never issued simply is not there.
That is the difference between a decorative certificate and a credential. Most certificates a student receives are unverifiable pieces of paper whose only claim to authenticity is a logo. Ours are designed to be interrogated, and the ability to interrogate them is published on the certificate itself.
Printable Papers, Rendered Correctly
Getting an exam paper to print as a true A4 page — from a phone — is harder than it looks. Here is how we solved it.
Browsers print what they see, and on a phone what they see is a layout built for a 390-pixel-wide screen. Printed naively, an examination paper produced on a phone comes out mis-scaled and unusable, which defeats the purpose of a printable paper. Rather than accept that, we render every printable document — examination papers, answer keys and certificates — inside an isolated hidden frame that is pinned to a fixed A4 viewport of 794 pixels, the exact width of 210mm at 96 DPI.
That frame carries its own print stylesheet and its own page geometry, so the document being printed is completely independent of the device that triggered the print. A paper generated on a budget Android phone and the same paper generated on a desktop produce identical, correctly proportioned A4 output. We verify this with automated browser tests at both mobile and desktop widths rather than trusting that it still works.
Examination papers themselves are assembled in your browser from a curated question bank, following the CBSE blueprint for section structure and marks. Because generation happens locally and pseudo-randomly, students can produce practice papers repeatedly without a server, and two students revising the same chapter do not receive an identical paper.
How We Know It Actually Works
Claims about quality are cheap. These are the checks we hold the site to.
- Zero horizontal overflow on every page at 375px — verified by automated browser tests, not by eye.
- Zero JavaScript errors across every page in the site, checked the same way.
- Semantic HTML with landmarks, a skip link, one
<h1>per page, and a logical heading order. - Keyboard operability with visible focus indicators, and interactive targets sized at least 44×44 pixels.
- Reduced-motion support: animation is disabled wherever the operating system asks for it.
- Progressive enhancement: content is readable when JavaScript fails, and reveal animations can never leave text permanently invisible.
- Valid structured data on every page, machine-checked for parse errors before release.
Built to Be Understood by Machines
Search engines and AI assistants are readers too, and we publish for them deliberately.
Structured Data
Every page carries Schema.org JSON-LD — organisation, breadcrumbs, courses, events and FAQs — so systems can interpret what a page is rather than guess.
sitemap.xml
A complete, dated sitemap listing every indexable page, referenced from robots.txt so crawlers can discover the whole site in one request.
llms.txt
A plain-language summary of what this organisation is and which pages matter, published for large language models and AI search systems.
security.txt
A machine-readable contact channel at /.well-known/security.txt, following RFC 9116, so a researcher can find the right address instantly.
Tell Us, and We Will Fix It
If something on this platform is broken, inaccessible, or insecure, we would genuinely rather hear about it than not. Accessibility barriers can be reported through our accessibility statement, security issues through our responsible disclosure policy, and anything else through Contact. Reporting a problem in good faith will never count against you.