Introduction to Low-Level Design (LLD)¶

1. What is Low-Level Design (LLD)?¶

Definition:
Designing the internal structure (“skeleton”) of an application by identifying: - Classes / objects
- Relationships
- Data flow
- How DSA solutions fit into the system

DSA vs LLD¶

DSA: Solves isolated problems
Examples: Binary search, Quicksort, Dijkstra’s, Heaps
LLD:
Decides what objects exist
Defines how they interact
Then applies DSA within this structure

2. Illustrative Story: QuickRide¶

Scenario: Build a ride-booking app like Uber/Ola

Anurag’s DSA-First Approach¶

Problem decomposition: - City → graph (nodes = intersections, edges = roads)
- Use Dijkstra’s → shortest path
- Use min-heap → match riders to closest drivers

Gaps: - No entities (User, Rider, Location, etc.)
- No data security (phone masking missing)
- No integrations (notifications, payments)
- No scalability planning

Maurya’s LLD-First Approach¶

1. Entity identification: - User
- Rider
- Location
- NotificationService
- PaymentGateway

2. Relationships & interactions: - User and Rider connected via Location
- Notification and Payment services integrated

3. Non-functional concerns: - Data security
- Scalability

4. Apply DSA: - Shortest path for routing
- Heap for driver matching

3. Core LLD Principles¶

Scalability¶

Handle large number of users
Easy to expand system

Maintainability¶

Easy to debug
Changes should not break existing features

Reusability¶

Loosely coupled modules
Example: reusable notification system

4. What LLD is NOT (vs HLD)¶

LLD does NOT include: - Tech stack (Java, Spring Boot, etc.)
- Database choice (SQL / NoSQL)
- Deployment (AWS, GCP, scaling)
- Cost optimization

5. Summary¶

DSA: Solves specific problems
LLD: Defines structure and interactions
HLD: Defines overall system architecture

6. Key Line¶

If DSA is the brain, LLD is the skeleton of your application.