📚 Study Tracker — ML, System Design & MLOps

A full-stack study tracker to systematically prepare for ML & System Design interviews. Built by @R1patil · Intern @ Raysoft.ai

🌐 Live Demo  |  📺 YouTube Channel  |  ⭐ Star this repo if you find it useful!

🏆 Proudly built for the Coral Hackathon

---

✨ Features

Feature	Description
🔐 Auth	Email + Password signup/login via Supabase
🧠 Jarvis AI Mentor	An empathetic, high-EQ AI agent that monitors your progress, calendar, and screen time using Coral SQL, guiding you with first-principles and yogic wisdom.
🕵️ Desktop Activity Tracker	A local Windows daemon that tracks your active windows and detects 5-minute "distraction spikes", pushing real-time interventions.
👤 Per-user data	Every user has their own private progress in Supabase PostgreSQL
✅ Topic tracking	Mark topics as Not Started / In Progress / Done
📝 Notes	Add personal notes and key takeaways per topic
⏱️ Study Timer	Per-topic and global session timer
🔥 Streaks	Daily study streak tracking
📊 Dashboard	Live progress % across all tracks
📺 YouTube Challenges	Add, embed, and track your own YouTube videos
🔗 Direct links	Every topic links directly to its resource

---

🐚 Powered by Coral (Hackathon Submission)

This project was built to showcase the incredible power of the Coral SQL engine for AI Agent tool-calling.

Instead of writing dozens of complex API wrappers to fetch GitHub repository data or internal PostgreSQL metrics, Jarvis (our LLaMA-3.3 70B AI Agent) is equipped with a single tool: execute_coral_sql.

By leveraging Coral, the AI can:

1. Query GitHub Natively: Analyze readme.md, directory structures, and branches via SELECT * FROM github.trees WHERE owner = '...'.
2. Monitor Student Behavior: Execute queries against student_activity.activity, student_progress.progress, and student_calendar.events to instantly know if you've been distracted by Instagram during your scheduled study blocks.
3. Zero Boilerplate: Coral eliminates the need for massive data extraction layers. The AI simply writes ANSI SQL, and Coral delivers the structured JSON results instantly.

---

🗺️ Learning Tracks

The curriculum is sourced from 3 of the most popular GitHub repos in the community:

🏗️ System Design

Source: awesome-system-design-resources

• Core Concepts (Scalability, CAP Theorem, Consistent Hashing...)
• Networking Fundamentals (OSI, DNS, Load Balancing...)
• API Fundamentals (REST, GraphQL, WebSockets...)
• Database Fundamentals (ACID, Sharding, Indexing...)
• Caching (Strategies, CDN, Eviction Policies...)
• Distributed Systems & Microservices
• Interview Problems (Easy → Medium → Hard)

🤖 Machine Learning

Source: machine-learning-interview

• ML Fundamentals (Linear Regression, SVM, XGBoost...)
• Deep Learning (CNNs, Transformers, Backpropagation...)
• ML System Design (Recommendation Systems, Fraud Detection...)
• ML Coding Questions (Implement from scratch)

⚙️ MLOps Zero to Hero

Source: mlops-zero-to-hero

• Containerization (Docker, Dockerfile, Docker Compose)
• Orchestration (Kubernetes, Helm, Kubeflow)
• CI/CD for ML (GitHub Actions, Jenkins)
• Experiment Tracking (MLflow, DVC, W&B)
• Model Serving & Monitoring (FastAPI, BentoML, Evidently)

---

🛠️ Tech Stack

Frontend

• Next.js 14 — App Router, Server Components
• TypeScript — Type-safe code
• Tailwind CSS — Styling
• Supabase SSR — Auth & session management

Backend

• FastAPI — REST API
• Coral CLI — SQL Engine for GitHub & PostgreSQL integrations
• Groq API — Ultra-fast LLaMA-3.3-70B inference for Jarvis AI
• Python 3.11 — Core language
• httpx — Async HTTP for Supabase queries and JWT verification

Infrastructure

• Supabase — Authentication + Row Level Security
• Vercel — Frontend hosting (free)
• Render — Backend hosting (free)
• GitHub — Source control + CI/CD

---

🚀 Getting Started Locally

Prerequisites

• Node.js 18+
• Python 3.11+
• A Supabase account (free)

1. Clone the repo

git clone https://github.com/R1patil/study-tracker.git
cd study-tracker

2. Setup Backend

cd backend
pip install -r requirements.txt

Create backend/.env:

SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=your_anon_key

Run:

uvicorn main:app --reload
# API → http://localhost:8000
# Docs → http://localhost:8000/docs

3. Setup Frontend

cd frontend
npm install

Create frontend/.env.local:

NEXT_PUBLIC_SUPABASE_URL=https://your-project.supabase.co
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_anon_key
NEXT_PUBLIC_API_URL=http://localhost:8000

Run:

npm run dev
# App → http://localhost:3000

4. Setup Supabase Database

Run this in your Supabase SQL Editor:

create extension if not exists "uuid-ossp";

create table user_progress (
  id uuid default uuid_generate_v4() primary key,
  user_id uuid references auth.users(id) on delete cascade not null,
  topic_id text not null,
  status text default 'not_started',
  notes text default '',
  updated_at timestamptz default now(),
  unique(user_id, topic_id)
);

create table study_sessions (
  id uuid default uuid_generate_v4() primary key,
  user_id uuid references auth.users(id) on delete cascade not null,
  topic_id text,
  topic_title text,
  duration_mins float default 0,
  date date default current_date,
  created_at timestamptz default now()
);

alter table user_progress enable row level security;
alter table study_sessions enable row level security;

create policy "Users own their progress" on user_progress
  for all using (auth.uid() = user_id);

create policy "Users own their sessions" on study_sessions
  for all using (auth.uid() = user_id);

---

📁 Project Structure

study-tracker/
├── backend/
│   ├── main.py              ← FastAPI app + JWT auth + all routes
│   ├── requirements.txt
│   ├── Dockerfile
│   └── .env                 ← (not committed)
│
└── frontend/
    ├── src/
    │   ├── app/
    │   │   ├── page.tsx              → Redirects to /login
    │   │   ├── login/page.tsx        → Login page
    │   │   ├── signup/page.tsx       → Signup page
    │   │   ├── dashboard/page.tsx    → Main dashboard
    │   │   ├── youtube/page.tsx      → YouTube challenges
    │   │   └── auth/callback/        → OAuth callback handler
    │   ├── components/
    │   │   ├── Sidebar.tsx
    │   │   ├── StatsHeader.tsx
    │   │   ├── TimerWidget.tsx
    │   │   ├── TrackView.tsx
    │   │   ├── TopicCard.tsx
    │   │   └── auth/UserMenu.tsx
    │   └── lib/
    │       ├── api.ts                ← API calls with JWT
    │       └── supabase/             ← Supabase client/server
    ├── .env.local            ← (not committed)
    └── package.json

---

🔌 API Endpoints

Method	Endpoint	Description
GET	/	Health check
GET	/progress	Get full user progress
GET	/stats	Get stats + streaks
PATCH	/topic/{id}/status	Update topic status
PATCH	/topic/{id}/notes	Update topic notes
GET	/timer	Get active timer
POST	/timer	Start / stop timer

All endpoints require Authorization: Bearer <supabase_jwt> header

---

☁️ Deployment — Free Forever

Service	Platform	Cost
Frontend	Vercel	✅ Free
Backend	Render	✅ Free
Auth + DB	Supabase	✅ Free

Deploy Backend → Render

1. Connect GitHub repo on render.com
2. Root Directory: backend
3. Add env vars: SUPABASE_URL, SUPABASE_ANON_KEY
4. Click Deploy

Deploy Frontend → Vercel

1. Import repo on vercel.com
2. Root Directory: frontend
3. Add env vars: NEXT_PUBLIC_SUPABASE_URL, NEXT_PUBLIC_SUPABASE_ANON_KEY, NEXT_PUBLIC_API_URL
4. Click Deploy

---

📖 How to Use

1. Sign up with your email → confirm via the email link
2. Pick a track from the sidebar — System Design, ML, or MLOps
3. Click the dot next to any topic to cycle status: Not Started → In Progress → Done
4. Click ⏱ to start a per-topic timer
5. Click 📝 to add personal notes and key takeaways
6. Hit ▶ Start Session at the top to log your study time
7. YouTube Challenges → paste any video URL → watch inline → mark as watched
8. Study every day to build your 🔥 streak!

---

🤝 Contributing

Pull requests are welcome! For major changes, please open an issue first to discuss what you would like to change.

---

📄 License

MIT — free to use and modify!

---

Made with ❤️ by Rahul Patil · Intern @ Raysoft.ai

⭐ Star this repo if it helped you prepare!