SHIVANSH AGARWAL
✦ CASE STUDY02 / 04
← Back to all work

CASE STUDY 02 — AI LEARNING PLATFORM

DainikBhaskar.ai

As part of the core team, I helped build the learning platform end to end — a custom LMS with courses, secure payments, daily live lectures and fast on-demand video. The backend runs on FastAPI, Fastify and Python with a React frontend, and was engineered to stay stable under heavy, high-concurrency traffic.

FASTAPIFASTIFYPYTHONREACTSQL · SQLALCHEMYCLERK AUTHRAZORPAYVIDEO STREAMINGLOAD BALANCINGLOAD TESTING
500+concurrent learners supported during peak usage
20%faster content-fetching & video-access flows
30%reduction in system error rate
10K+requests/sec target, load-tested behind a load balancer

OVERVIEW

A learning platform built to scale — from courses to live classes.

I designed and developed backend APIs and workflows for users, courses, payments, videos, live lectures, recordings and learning-content management. The goal was a platform that stays fast and dependable whether a learner is browsing a course, checking out securely, or joining a live class alongside thousands of others.

Much of the work was building from scratch on the core team: shaping the LMS, integrating third-party services, hardening the backend, and turning ideas into working demos to show investors how the product should behave.

WHAT I BUILT

The pieces behind the platform.

LMS

Custom learning system

Courses, lessons, recordings and learning-content management — built from the ground up rather than bolted onto an off-the-shelf tool.

AUTH

Clerk authentication

Secure sign-up and sign-in with Clerk, wired into the API so every user, course and payment action is properly authenticated.

PAYMENTS

Razorpay integration

Integrated Razorpay for secure checkout and course payments, with clean transaction flows and reliable records.

MEDIA

Fast video delivery

Videos are hosted and served through an optimized delivery path, so playback starts quickly with minimal delay across live and recorded sessions.

LIVE

Live lectures & recordings

Daily live classes plus recordings, made available in the platform so learners can join live or catch up later.

SCALE

Load-tested scalability

APIs put behind a load balancer and load-tested for heavy concurrent traffic while keeping response times low and stable.

ENGINEERING HIGHLIGHTS

Optimized for speed and reliability.

BEYOND THE LMS

Demos, workflows and a second platform.

Alongside the learning platform, I built product demos and workflows from scratch to show investors how features should work — translating early product ideas into interactive, working prototypes.

I also contributed to BBV (Bharat Bhagya Vidhata), a second LMS, applying the same backend and platform experience to a different learning product.

TECH STACK

What it's built with.

Backend

FastAPI, Fastify, Python — REST APIs and workflows

Frontend

React

Data

SQL with SQLAlchemy; careful schema and data management

Auth

Clerk authentication

Payments

Razorpay integration

Media

Hosted video with optimized delivery for live & recorded lectures

Scale

Load balancing and load testing for high concurrency

A NOTE

This project was built inside the company as part of the core team. Some product internals and additional work are proprietary and intentionally left out here — happy to talk through more of what I did in a conversation.

See it live.