🎓 CampusOS — Autonomous Multi-Agent Campus AI System

StudAI Foundry 2026 · SRMIST Kattankulathur Track 1: Autonomous Campus & Community Systems

---

🧠 What is CampusOS?

CampusOS is a fully autonomous multi-agent AI system that replaces the fragmented campus experience at SRMIST. Students interact in natural language — the system thinks, plans, executes, reviews, and self-corrects autonomously.

Instead of navigating 6+ portals, students just say:

"Register me for all data science events, remind me 1 day before each, and find research papers on those topics"

CampusOS handles all 3 actions simultaneously with full autonomy trace.

---

⚡ Autonomy Loop (Think → Plan → Execute → Review → Update)

Student Query (natural language / Hindi / Tamil)
        ↓
🧠 THINK   — GPT-4o + regex intent classifier (94.2% accuracy)
        ↓
📋 PLAN    — TaskPlanner generates ordered execution steps
        ↓
⚡ EXECUTE — 7 specialist agents with retry logic (max 2 retries)
        ↓
✅ REVIEW  — Deep validation per intent (DB verification, time checks)
        ↓
🔄 UPDATE  — Self-correction + dynamic replanning on failure

---

🤖 7 Specialist Agents

Agent	Handles	Tool Used
🎓 StudentSupportAgent	Campus Q&A, attendance, rules	FAISS RAG (86 chunks)
🎫 EventOperationsAgent	Event discovery + registration	SQLite DB
🔬 LabAssistanceAgent	Equipment reservation + return	SQLite DB
📋 GrievanceRoutingAgent	Complaint filing + routing	SQLite DB
⏰ ReminderAgent	Smart reminders with AM/PM parsing	APScheduler
📄 DocumentGeneratorAgent	10 document templates + PDF	PDF Generator
📚 LibraryResearchAgent	arXiv papers + study notes	arXiv API

---

🏗️ Architecture

frontend/          React + Vite + TypeScript + Tailwind (port 8080)
backend/
  app/
    agents/        7 specialist agents + master orchestrator
    api/           FastAPI REST endpoints (20+)
    core/          LLM manager, vector store, scheduler, task planner
    models/        SQLAlchemy models (Events, Lab, Grievances, etc.)
    utils/         NLP patterns, notifications, language detection
  data/
    scraped/       srmist_data.json (knowledge base)
    generated/     Generated PDF documents
    faiss_index.pkl  Cached FAISS vector index

---

🛠️ Tech Stack

Layer	Technology
Frontend	React 18 + Vite + TypeScript + Tailwind + shadcn/ui
Backend	FastAPI + Python 3.11
LLM Primary	OpenAI GPT-4o
LLM Fallback	Groq Llama 3.3 (4 keys, round-robin)
Vector Store	FAISS + sentence-transformers (all-MiniLM-L6-v2)
Database	SQLite (SQLAlchemy ORM)
Scheduler	APScheduler (IST timezone)
PDF Generation	ReportLab + HTML fallback
Research	arXiv API

---

🚀 Quick Start

Prerequisites

• Python 3.11+
• Node.js 18+
• OpenAI API key
• Groq API keys (optional fallback)

Backend Setup

cd backend
python -m venv venv
venv\Scripts\activate          # Windows
# source venv/bin/activate     # Mac/Linux

pip install -r requirements.txt

# Create .env file
cp .env.example .env
# Add your API keys to .env

# Seed database with demo data
python scripts/seed_database.py

# Start backend
uvicorn app.main:app --reload --host 0.0.0.0 --port 8000

Frontend Setup

cd frontend
npm install
npm run dev
# Runs at http://localhost:8080

Demo Accounts

Name	Register Number	Department
Shri Harsan	RA2512052010001	Computing Technologies
Divya	RA2512052010002	Data Science & Business Systems
Rohith	RA2512052010003	Computing Technologies
Ishitha	RA2512052010004	Computing Technologies

---

🔧 Environment Variables

# backend/.env
OPENAI_API_KEY=sk-...
GROQ_API_KEY_1=gsk_...
GROQ_API_KEY_2=gsk_...
GROQ_API_KEY_3=gsk_...
GROQ_API_KEY_4=gsk_...
DATABASE_URL=sqlite:///campusos.db

---

🐳 Docker

docker-compose up --build
# Frontend: http://localhost:8080
# Backend:  http://localhost:8000
# API Docs: http://localhost:8000/docs

---

📊 Key Features

• Multi-agent chain — Register + Remind + Research in 1 query
• Self-correction — Already registered? Shows alternatives automatically
• RAG knowledge — 86 SRMIST-specific knowledge chunks
• Multilingual — English, Hindi, Tamil queries supported
• 10 document types — Leave letter, NOC, Bonafide, Study schedule, etc.
• Analytics dashboard — Agent usage, autonomy metrics, tool integrations
• Proactive suggestions — After booking IoT workshop → suggests Arduino reservation

---

👥 Team

Member	Role
Shri Harsan	Team Lead + AI Architecture
Divya	AI/ML + Agent Design
Rohith	Backend + Database
Ishitha	Frontend + UI/UX

---

🏆 Competition

• Event: StudAI Foundry 2026 @ SRMIST
• Track: Track 1 — Autonomous Campus & Community Systems
• Prize: ₹1,00,000
• Demo: March 26-28, 2026

---

📄 License

MIT License — see LICENSE