Building practical learning platforms, developer tools, and AI-enabled educational experiences.
Vastavik Coding Stuff is a GitHub organization dedicated to building educational technology for students and learners. Our core product is the Vastavik Learning ecosystem — a cross-platform suite of applications that provides interactive courses, live classrooms, AI-powered tutoring, hands-on code practice, and real-time collaboration tools.
We build for India-first learners — accessible, affordable, and practical. Every project in this organization is designed to help students learn by doing, not just watching.
- Educational platforms — Structured learning paths with courses, lessons, quizzes, and progress tracking
- Android & mobile applications — Native Kotlin apps with Jetpack Compose and Material 3
- Modern web experiences — SEO-optimized, performant Next.js applications
- Backend APIs & services — FastAPI-powered REST, WebSocket, and SSE backends
- AI-powered tutoring — Integration with Mistral AI and Google Gemini for contextual learning assistance
- Real-time collaboration — Live classrooms, WebRTC signaling, peer chat, and interactive whiteboards
- Programming tools — In-browser code execution, OCR exercise scanning, and practice environments
graph LR
A[Students / Learners] --> B[Android App]
A --> C[Web App]
B --> D[Backend Services]
C --> D
D --> E[AI Tutoring]
D --> F[Code Execution]
D --> G[Real-Time Signaling]
The Vastavik Learning ecosystem consists of three primary applications that share a common backend. Students interact through either the Android app or the web app, while the backend services handle authentication, course management, AI tutoring, code execution, payments, and real-time communication.
| Repository | vastavikLearning-backend-app |
| Language | Python 3.11+ |
| Framework | FastAPI (async) |
| Database | Google Firebase Firestore + in-memory fallback |
| License | Apache License 2.0 |
The backend service layer powering the entire Vastavik Learning platform. Built with FastAPI and Uvicorn, it provides REST APIs, WebSocket real-time channels, and Server-Sent Events for AI streaming.
Key capabilities:
- Authentication — SHA-256 + salt password hashing, JWT (HS256) access/refresh tokens, HMAC-SHA256 request signing, Google OAuth, GitHub OAuth
- Real-time — Socket.IO bidirectional events, native WebSockets for peer chat, WebRTC SDP/ICE signaling relay
- AI Tutoring — Mistral AI (primary) with Google Gemini fallback chain, local pedagogical fallback for offline scenarios
- Code Execution — Judge0 self-hosted sandbox supporting Python, JavaScript, Java, C++, C, and SQL
- Payments — PhonePe, Razorpay, Cashfree webhook integration
- Security — Dynamic route circuit breaker, in-memory token bucket rate limiter, NGINX reverse proxy with HTTP/2 and SSL
- Deployment — Self-hosted Ubuntu VPS with systemd + NGINX + UFW + Fail2ban, Docker Compose, Railway, Render
| Repository | vastavikLearning-app |
| Language | Kotlin 2.0.20 |
| UI | Jetpack Compose (100%) + Material 3 |
| Architecture | MVVM + Clean Architecture |
| License | Apache License 2.0 |
Native Android client for student-focused learning experiences. Built entirely with Jetpack Compose — zero XML layouts — featuring a custom Neo-Brutalist design system with bold outlines, hard drop shadows, and high-contrast palettes.
Key capabilities:
- 11 ViewModels — Auth, Home, Learning, Chat, Meeting, Practice, Quiz, VideoLesson, Onboarding, Settings, Profile
- 30+ Navigation Routes — Full app navigation with deep linking via Firebase Dynamic Links
- Firebase Suite — Auth, Firestore, Storage, Cloud Messaging, Analytics (BOM 33.7.0)
- AI Integration — Mistral AI (3 models), Google Gemini (2 models), Google Generative AI SDK, MLKit Text Recognition for OCR exercise scanning
- Media — YouTube Player (embedded playback), Coil (image loading), Lottie (animations), CameraX (OCR capture)
- Real-time — OkHttp WebSockets for peer chat, WebRTC signaling, SSE streaming for AI responses
- Security — HMAC-SHA256 request signing, JWT auto-refresh via OkHttp Authenticator, device root detection, FLAG_SECURE
- Background — WorkManager periodic engagement notifications, Foreground Service for live meetings, in-app OTA updates via GitHub Releases
- Testing — JUnit 4, Espresso, Compose UI Test
| Repository | vastavikLearning-web |
| Language | TypeScript 5.6.3 (strict) |
| Framework | Next.js 14.2.35 (App Router) |
| CSS | Tailwind CSS 3.4.13 + custom Neo-Brutalist design system |
| Live Site | vastaviklearning.online |
| License | Apache License 2.0 |
Web experience for the Vastavik Learning platform. A deliberately minimal, zero-dependency Next.js application with only 3 runtime dependencies (next, react, react-dom). Every UI component, the auth system, and the entire design system are hand-built from scratch.
Key capabilities:
- 27 Routes — Complete web application covering courses, lessons, practice, quiz, live classrooms, AI chat, whiteboard, leaderboard, PYQ archive, and pricing
- SEO — Dynamic sitemap, robots.txt, Schema.org JSON-LD structured data (Organization, Course, FAQPage, Product), Open Graph, Twitter Cards, canonical URLs, PWA manifest
- Security — 12+ HTTP security headers (HSTS, CSP, X-Frame-Options, COOP/COEP/CORP), documented threat model and security policy
- Payments — Razorpay UPI Autopay integration with cards, netbanking, and wallet support
- Fonts — Space Grotesk (display), Inter (body), JetBrains Mono (code) via Google Fonts
- Custom Design System — 580-line CSS with 6-color accent palette (Yellow, Pink, Blue, Lime, Orange, Purple), hard offset shadows, thick borders, bold typography
- Deployment — Vercel with SSG (Static Site Generation) via
generateStaticParams()
| Category | Technologies |
|---|---|
| Languages | Python, Kotlin, TypeScript |
| Mobile | Android (Kotlin), Jetpack Compose, Material 3 |
| Web | Next.js 14, React 18, TypeScript, Tailwind CSS |
| Backend | FastAPI, Uvicorn, Pydantic v2 |
| Database | Google Firebase Firestore |
| Real-time | Socket.IO, WebSockets, WebRTC, Server-Sent Events |
| AI/ML | Mistral AI, Google Gemini, Google Generative AI SDK, MLKit |
| Authentication | JWT (HS256), HMAC-SHA256, Google OAuth, GitHub OAuth |
| Payments | Razorpay (UPI Autopay), PhonePe, Cashfree |
| Code Execution | Judge0 (self-hosted) |
| Containerization | Docker, Docker Compose |
| Reverse Proxy | NGINX (HTTP/2, SSL, rate limiting) |
| Deployment | Vercel, Railway, Render, Self-hosted Ubuntu VPS |
| Tooling | Git, GitHub, Gradle, PostCSS |
| Repository | Description | Primary Language | Link |
|---|---|---|---|
| vastavikLearning-backend-app | Backend services and API layer for the Vastavik Learning platform | Python | View → |
| vastavikLearning-app | Native Android client for student learning experiences | Kotlin | View → |
| vastavikLearning-web | Web experience for the Vastavik Learning platform | TypeScript | View → · Live → |
| .github | Organization profile, README, and shared resources | — | View → |
- Learner-first design — Every feature starts with the question: "Does this help a student learn better?"
- Maintainable code — Clean architecture, consistent patterns, and thorough documentation across all repositories
- Iterative development — Ship early, gather feedback, improve continuously. Every project is actively evolving
- Open collaboration — We welcome issues, feature requests, and pull requests from the community
- Responsible AI — AI features are designed as learning assistants, not replacements for understanding. Human judgment remains central
All three application repositories (backend, Android app, web app) are released under the Apache License 2.0.
To contribute:
- Explore the repositories and find something you'd like to improve
- Open an issue in the relevant repository to discuss your proposal
- Fork and branch — create a feature branch from
main - Submit a pull request — describe your changes clearly and link any related issues
Check each repository's LICENSE file for the authoritative license terms.
Learning through building. Building for learners.