Ride sharing application Backend

RideFlow is a scalable, production-ready backend for a ride-sharing platform built with FastAPI and Redis. It implements rate-limiting, asynchronous FIFO ride queueing, and parallel worker-based driver assignment, demonstrating modern backend engineering and distributed systems design.

---

Features

• FastAPI REST API
  Modern, type-safe endpoints for ride requests and management.

• Redis-backed FIFO queue
  Ensures rides are processed in order and allows decoupling of API and worker processes.

• Async Worker Pool
  Multiple workers process queued rides in parallel for real-time driver assignment.

• Rate Limiting
  Per-client IP rate limiting to prevent abuse.

• Non-blocking Architecture
  API endpoints remain responsive even under heavy load.

• Clean Project Structure
  Absolute imports, modular architecture, and best practices for maintainable backend code.

---

Technologies Used

Layer	Technology
Backend Framework	FastAPI
Queue & Caching	Redis
Async Workers	Python asyncio + Redis
Data Format	JSON
Server	Uvicorn ASGI

---

Project Structure

ride-sharing-backend/
├── app/
│   ├── __init__.py
│   ├── main.py              # FastAPI application
│   ├── ride_queue.py        # Redis-based enqueue/dequeue
│   ├── workers.py           # Async worker logic
│   ├── worker_runner.py     # Worker entrypoint
│   ├── matcher.py           # Driver matching logic
│   ├── rate_limiter.py      # Redis-based rate limiting
│   └── schemas.py           # Pydantic request/response schemas
├── .venv/                   # Python virtual environment
└── requirements.txt

---

Quick Start

Install Dependencies

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Start Redis

redis-server

Run FastAPI Server

uvicorn app.main:app --reload

Start Worker Pool (Separate Terminal)

python app/worker_runner.py

Access Swagger UI

Open in browser:

http://127.0.0.1:8000/docs

---

Testing

Sample Ride Request

POST /request-ride with JSON:

{
  "rider_id": 1,
  "pickup_lat": 12.97,
  "pickup_lng": 77.59,
  "dest_lat": 13.01,
  "dest_lng": 77.60
}

Expected outcome:

• API responds with "Ride queued successfully"
• Worker terminal prints driver assignment:

Worker 0 → Driver 4821 assigned

---

Architecture Overview

1. API receives ride requests → validated with Pydantic schemas.
2. Rate limiter checks client IP → prevents abuse.
3. Ride queued in Redis FIFO list → decouples API and processing.
4. Async worker pool continuously dequeues rides → assigns available drivers.
5. Drivers assigned → response printed/logged (can be extended to WebSockets or DB).

Diagram:

[Rider Request] → [FastAPI API] → [Redis Queue] → [Async Workers] → [Driver Assignment]

---

Dependencies

fastapi
uvicorn[standard]
redis
pydantic

---