Design Food Delivery App

A food delivery app enables users to browse restaurants, place orders, and get meals delivered to their doorstep with features like real-time tracking, multiple payment options.

Shashank Mishra

Jan 04, 2025

1.Users must be able to sign up/login using email, phone number, or social media.

2.User should be able to onboard restaurant

3.Search restaurants by name, cuisine, or specific dishes.

4.Register and authenticate delivery agents.

5.System should be able to view delivery tasks assigned to them.

6.Notify delivery agents about new delivery requests.

7.Allow customers to track the status of their order (order placed, accepted, being prepared, out for delivery, delivered).

8.Notify customers about order status, special offers, and restaurant updates.

9.Allow customers to rate and review

10.Suppor multiple payment system

NFR

1.The system should be highly available( 99.99% of the time) to handle orders without downtime.

2.Search results for restaurants should load within 2 seconds. and Order placement and payment processing should complete within 3-5 seconds(low latency system)

3.System should be highly scalable to support thousands of concurrent users, restaurants, and delivery agents.

4.System should be strongly consistent

5.System should be fault tolerant

Estimates

Total users=1B
DAU=100M users
Each user makes 5 searches per day
total searches =100M*5=500M
Search QPS=500M/10^5=5k QPS
For peak = 5k*10=50k QPS

For write  assume only 10% place orders=100M*10/100*5(per day 5 orders)=50M
write QPS=50M/10^5=500 QPS

Read:write=10:1

Storage

For users table =1B*1kB
For orders table = 50M*1KB=50 GB
for 1 years=50GB*400(~365) =20000GB=20TB
For 10 years=20TB*10=200TB
Number of restaurants= 100k*1MB(image for s3)=100GB
for metadata = 100k *100KB=10GB

API’s

POST /api/v1/users/register

request {
  "name": "John Doe",
  "email": "johndoe@example.com",
  "phone": "+1234567890",
  "password": "securepassword123"
}
response {
  "status": "success",
  "message": "User registered successfully",
  "user_id": "12345"
}

2.POST /api/v1/users/login

request {
  "email": "johndoe@example.com",
  "password": "securepassword123"
}
response {
  "status": "success",
  "message": "Login successful",
  "auth_token": "abcdef1234567890",
  "user_profile": {
    "user_id": "12345",
    "name": "John Doe",
    "email": "johndoe@example.com",
    "phone": "+1234567890",
    "addresses": ["123 Street, City, Country"]
  }
}

3.PUT /api/v1/users/{userId}

request {
  "name": "John Updated",
  "email": "johnupdated@example.com",
  "addresses": ["456 Avenue, City, Country"],
  "phone": "+9876543210"
}
response {
  "status": "success",
  "message": "Profile updated successfully",
  "updated_profile": {
    "user_id": "12345",
    "name": "John Updated",
    "email": "johnupdated@example.com",
    "phone": "+9876543210",
    "addresses": ["456 Avenue, City, Country"]
  }
}

4.GET /api/v1/users/{userId}

{
  "user_id": "12345",
  "name": "John Updated",
  "email": "johnupdated@example.com",
  "phone": "+9876543210",
  "addresses": ["456 Avenue, City, Country"],
  "payment_methods": [
    {"type": "Credit Card", "last_four": "1234"},
    {"type": "PayPal", "email": "johnupdatedpaypal@example.com"}
  ]
}

5.POST /api/v1/restaurants/register

request {
  "restaurant_name": "Food Paradise",
  "location": "123 Restaurant Street, City, Country",
  "contact_email": "contact@foodparadise.com",
  "contact_phone": "+9876543210",
  "menu": [
    {"name": "Pizza", "price": 12.99, "description": "Cheese pizza with toppings"},
    {"name": "Pasta", "price": 10.99, "description": "Spaghetti with tomato sauce"}
  ]
}

response {
  "status": "success",
  "message": "Restaurant registered successfully",
  "restaurant_id": "7890"
}

6.POST /api/v1/restaurants/login

request {
  "email": "admin@foodparadise.com",
  "password": "adminsecurepassword"
}
 response {
  "status": "success",
  "message": "Login successful",
  "auth_token": "abcdef9876543210",
  "restaurant_profile": {
    "restaurant_id": "7890",
    "restaurant_name": "Food Paradise",
    "contact_email": "contact@foodparadise.com",
    "contact_phone": "+9876543210",
    "menu": [
      {"name": "Pizza", "price": 12.99, "description": "Cheese pizza with toppings"},
      {"name": "Pasta", "price": 10.99, "description": "Spaghetti with tomato sauce"}
    ]
  }
}

7.PUT /api/v1/restaurants/{restaurantId}/menu

request {
  "menu": [
    {"name": "Burger", "price": 8.99, "description": "Beef burger with cheese"},
    {"name": "Salad", "price": 5.99, "description": "Mixed greens with dressing"}
  ]
}
response {
  "status": "success",
  "message": "Menu updated successfully",
  "updated_menu": [
    {"name": "Burger", "price": 8.99, "description": "Beef burger with cheese"},
    {"name": "Salad", "price": 5.99, "description": "Mixed greens with dressing"}
  ]
}

8.POST /api/v1/delivery-partners/register

request {
  "name": "Sarah Conner",
  "phone": "+9876543211",
  "vehicle_type": "Bike",
  "vehicle_number": "AB123CD"
}
response {
  "status": "success",
  "message": "Delivery partner registered successfully",
  "delivery_partner_id": "4567"
}

9.POST /api/v1/delivery-partners/login

request {
  "phone": "+9876543211",
  "password": "deliverypassword"
}
resposne {
  "status": "success",
  "message": "Login successful",
  "auth_token": "abcdef6543210",
  "delivery_partner_profile": {
    "delivery_partner_id": "4567",
    "name": "Sarah Conner",
    "phone": "+9876543211",
    "vehicle": {
      "type": "Bike",
      "number": "AB123CD"
    },
    "assigned_orders": [
      {"order_id": "112233", "status": "Out for delivery", "pickup_location": "Restaurant XYZ"}
    ]
  }
}

10.POST /api/v1/orders

request {
  "user_id": "12345",
  "restaurant_id": "7890",
  "items": [
    {"menu_item_id": "1", "quantity": 2},
    {"menu_item_id": "2", "quantity": 1}
  ],
  "delivery_address": "456 Avenue, City, Country",
  "payment_method": "Credit Card",
  "payment_details": {"card_number": "4111111111111111", "expiry": "12/23", "cvv": "123"}
}
response {
  "status": "success",
  "message": "Order placed successfully",
  "order_id": "98765",
  "order_status": "Pending"
}

11.PUT /api/v1/orders/{orderId}/status

request {
  "status": "Out for delivery"
}
response {
  "status": "success",
  "message": "Order status updated successfully",
  "order": {
    "order_id": "98765",
    "status": "Out for delivery",
    "assigned_delivery_partner": "Sarah Conner"
  }
}

12.GET /api/v1/orders/{orderId}

response {
  "order_id": "98765",
  "status": "Out for delivery",
  "items": [
    {"menu_item_name": "Pizza", "quantity": 2, "price": 12.99},
    {"menu_item_name": "Pasta", "quantity": 1, "price": 10.99}
  ],
  "total_price": 36.97,
  "delivery_address": "456 Avenue, City, Country",
  "assigned_delivery_partner": "Sarah Conner"
}

13.POST /api/v1/reviews

request {
  "user_id": "12345",
  "restaurant_id": "7890",
  "rating": 4,
  "review": "Great food, quick delivery!"
}
resposne {
  "status": "success",
  "message": "Review added successfully",
  "review_id": "5678"
}

14.POST /api/v1/payments/initiate

request {
  "order_id": "98765",
  "user_id": "12345",
  "amount": 36.97,
  "payment_method": "Credit Card",
  "payment_details": {
    "card_number": "4111111111111111",
    "expiry": "12/23",
    "cvv": "123"
  }
}
response {
  "status": "success",
  "message": "Payment initiated successfully",
  "payment_id": "abcd1234",
  "payment_status": "pending"
}

GET /api/v1/payments/{paymentId}

{
  "payment_id": "abcd1234",
  "order_id": "98765",
  "payment_status": "completed",
  "payment_method": "Credit Card",
  "transaction_id": "txn_5678",
  "amount": 36.97,
  "currency": "USD"
}

Databases

1. Relational Databases (RDBMS)

Use Case: Managing structured and transactional data (e.g., user profiles, orders, payments).
DB Options: MySQL, PostgreSQL, Amazon RDS.

Schema Design

1. Users Table

CREATE TABLE Users (
    user_id UUID PRIMARY KEY,
    name VARCHAR(255),
    email VARCHAR(255) UNIQUE,
    phone VARCHAR(15) UNIQUE,
    password_hash VARCHAR(255),
    addresses JSONB, -- Multiple addresses in JSON format
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP
);

2. Restaurants Table

CREATE TABLE Restaurants (
    restaurant_id UUID PRIMARY KEY,
    name VARCHAR(255),
    location VARCHAR(255),
    contact_email VARCHAR(255),
    contact_phone VARCHAR(15),
    menu JSONB, -- Stores menu details
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

3. Orders Table

CREATE TABLE Orders (
    order_id UUID PRIMARY KEY,
    user_id UUID REFERENCES Users(user_id),
    restaurant_id UUID REFERENCES Restaurants(restaurant_id),
    total_amount DECIMAL(10, 2),
    order_status ENUM('Pending', 'Confirmed', 'Out for Delivery', 'Delivered', 'Cancelled'),
    delivery_address TEXT,
    payment_id UUID,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

4. Payments Table

CREATE TABLE Payments (
    payment_id UUID PRIMARY KEY,
    order_id UUID REFERENCES Orders(order_id),
    user_id UUID REFERENCES Users(user_id),
    amount DECIMAL(10, 2),
    payment_status ENUM('Pending', 'Completed', 'Failed', 'Refunded'),
    payment_method ENUM('Credit Card', 'UPI', 'Net Banking', 'Wallet'),
    transaction_id VARCHAR(255),
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

2. NoSQL Databases

Use Case: Storing unstructured or semi-structured data and real-time operations (e.g., menu updates, user sessions).
DB Options: MongoDB, DynamoDB, Couchbase.

Schema Design

1. Menu Collection (MongoDB)

{
    "restaurant_id": "7890",
    "menu": [
        {
            "menu_item_id": "1",
            "name": "Pizza",
            "price": 12.99,
            "description": "Cheese pizza with toppings",
            "availability": true
        },
        {
            "menu_item_id": "2",
            "name": "Pasta",
            "price": 10.99,
            "description": "Spaghetti with tomato sauce",
            "availability": true
        }
    ]
}

2. User Sessions Collection

{
    "user_id": "12345",
    "session_token": "abcdef1234567890",
    "created_at": "2025-01-03T10:30:00Z",
    "expires_at": "2025-01-03T12:30:00Z"
}

3. Real-Time Orders (Redis for caching)

Key-Value structure:

Order:98765
{
    "order_id": "98765",
    "user_id": "12345",
    "status": "Out for delivery",
    "assigned_delivery_partner": "Sarah Conner",
    "restaurant_name": "Food Paradise"
}

3. Search Databases

Use Case: Search functionalities like finding restaurants, menu items, or orders based on keywords or locations.
DB Options: Elasticsearch, Apache Solr.

Schema Design

Restaurant Index (Elasticsearch)

{
    "restaurant_id": "7890",
    "name": "Food Paradise",
    "location": {
        "latitude": 40.7128,
        "longitude": -74.0060
    },
    "menu_items": ["Pizza", "Pasta", "Burger", "Salad"],
    "rating": 4.5
}

Menu Items Index

{
    "menu_item_id": "1",
    "restaurant_id": "7890",
    "name": "Pizza",
    "price": 12.99,
    "description": "Cheese pizza with toppings",
    "tags": ["cheese", "toppings", "vegetarian"]
}

4. Analytics Databases

Use Case: Tracking trends, user behavior, and order analytics.
DB Options: Apache Cassandra, Google BigQuery, Snowflake, Amazon Redshift.

Schema Design

Order Analytics

Partition Key: restaurant_id
Clustering Key: order_date

CREATE TABLE OrderAnalytics (
    restaurant_id UUID,
    order_date DATE,
    total_orders INT,
    total_revenue DECIMAL(10, 2),
    PRIMARY KEY (restaurant_id, order_date)
);

User Activity Logs (Cassandra)

Partition Key: user_id
Clustering Key: activity_timestamp

CREATE TABLE UserActivityLogs (
    user_id UUID,
    activity_timestamp TIMESTAMP,
    activity_type TEXT,
    metadata MAP<TEXT, TEXT>, -- E.g., {"order_id": "98765"}
    PRIMARY KEY (user_id, activity_timestamp)
);

5. Time-Series Databases

Use Case: Tracking delivery partner locations or performance metrics over time.
DB Options: InfluxDB, TimescaleDB.

Schema Design

Delivery Partner Location Tracking

CREATE TABLE DeliveryLocations (
    delivery_partner_id UUID,
    timestamp TIMESTAMPTZ,
    latitude DECIMAL(9, 6),
    longitude DECIMAL(9, 6),
    PRIMARY KEY (delivery_partner_id, timestamp)
);

6. Graph Databases

Use Case: Capturing relationships between entities (e.g., users, orders, restaurants, delivery partners).
DB Options: Neo4j, Amazon Neptune.

Schema Design

Graph Schema

Nodes: Users, Restaurants, Delivery Partners, Orders
Edges: "PLACED" (User → Order), "DELIVERED_BY" (Order → Delivery Partner), "FULFILLED_BY" (Order → Restaurant)

Example Query (Neo4j): Find all orders delivered by a specific delivery partner.

MATCH (dp:DeliveryPartner)-[:DELIVERED_BY]->(o:Order)
WHERE dp.delivery_partner_id = "4567"
RETURN o;

7. Message Queues for Event-Driven Architecture

Use Case: Real-time communication for order status updates, notifications, and asynchronous processing.
Options: Apache Kafka, RabbitMQ, Amazon SQS.

Schema for Kafka Topics

Topic: order_events

{
  "event_type": "OrderPlaced",
  "order_id": "98765",
  "user_id": "12345",
  "restaurant_id": "7890",
  "timestamp": "2025-01-03T10:30:00Z"
}

Topic: delivery_updates

{
  "event_type": "OrderOutForDelivery",
  "order_id": "98765",
  "delivery_partner_id": "4567",
  "timestamp": "2025-01-03T11:00:00Z"
}

1. Caching

Redis is ideal for caching frequently accessed data to reduce database load and improve response times.

Examples:

Restaurant Menus: Cache menu details of popular restaurants.

Key: "restaurant_menu:12345"
Value: {
  "menu_items": [
    {"id": "1", "name": "Pizza", "price": 12.99},
    {"id": "2", "name": "Burger", "price": 8.99}
  ]
}

HLD

Shashank’s Substack

Discussion about this post