SDSystem Design Academy
A field guide for systems that survive the follow-up
60 production systems · continuously deployable

Design the system.Defend every choice.

An animated, long-form curriculum for learning why an architecture works, what alternatives were rejected, and how invariants, bottlenecks, failures, and cost shape each choice.

0deep dives
0thousand words
0architecture diagrams
LIVE ARCHITECTURE · REQUEST FLOW
packets flowing · no dependencies
Recommended path

Build the mental model in four moves.

Start with a compact read-heavy service, add distributed enforcement, learn ordered logs, then finish with a correctness-critical transaction system.

The curriculum

60 systems. No hand-waving.

Filter by architecture family or search for a concept. Every design explains why the architecture was selected, rejected alternatives, trade-offs, evolution triggers, failure analysis, and answer guidance.

60 of 60 chapters
01 / 60

URL Shortener

Create compact links, resolve them globally, and count clicks without slowing redirects.

Foundations3.5K+ words · rationale · 2 diagrams
02 / 60

Rate Limiter

Protect APIs with predictable quotas across a distributed fleet.

Foundations3.5K+ words · rationale · 2 diagrams
03 / 60

Notification System

Deliver push, email, and SMS reliably while respecting preferences and provider limits.

Product Systems3.5K+ words · rationale · 2 diagrams
04 / 60

File Storage & Sharing

Upload, sync, version, and share large files with durable metadata and blob storage.

Storage3.5K+ words · rationale · 2 diagrams
05 / 60

Chat & Messaging

Deliver ordered messages, receipts, presence, and offline sync at global scale.

Realtime3.5K+ words · rationale · 2 diagrams
06 / 60

News Feed

Generate ranked, paginated feeds while absorbing celebrity fan-out.

Product Systems3.5K+ words · rationale · 2 diagrams
07 / 60

Social Post Service

Create posts, media, follows, and public timelines with clear consistency boundaries.

Product Systems3.5K+ words · rationale · 2 diagrams
08 / 60

Video Streaming Platform

Ingest videos, transcode variants, and stream globally with adaptive bitrate playback.

Media3.5K+ words · rationale · 2 diagrams
09 / 60

Ride-Hailing Service

Track drivers, match nearby supply, and manage trips as a consistent state machine.

Geo & Realtime3.5K+ words · rationale · 2 diagrams
10 / 60

Ticket Booking

Reserve scarce seats under extreme contention without double booking.

Transactions3.5K+ words · rationale · 2 diagrams
11 / 60

API Gateway

Authenticate, route, protect, and observe traffic without becoming a single failure domain.

Infrastructure3.5K+ words · rationale · 2 diagrams
12 / 60

Distributed Cache

Serve low-latency reads with partitioning, eviction, replication, and predictable failure behavior.

Infrastructure3.5K+ words · rationale · 2 diagrams
13 / 60

Job Scheduler

Run delayed, recurring, and retryable work exactly when business semantics require.

Infrastructure3.5K+ words · rationale · 2 diagrams
14 / 60

Web Crawler

Fetch the web politely, deduplicate URLs, and scale parsing and storage independently.

Data Platforms3.5K+ words · rationale · 2 diagrams
15 / 60

Search Autocomplete

Return relevant prefix suggestions in milliseconds while trends change continuously.

Search3.5K+ words · rationale · 2 diagrams
16 / 60

Logging & Monitoring

Ingest, index, retain, query, and alert on telemetry under bursty load.

Observability3.5K+ words · rationale · 2 diagrams
17 / 60

Metrics Collection

Collect high-cardinality time series, aggregate them, and evaluate alerts reliably.

Observability3.5K+ words · rationale · 2 diagrams
18 / 60

Distributed Key-Value Store

Partition, replicate, and reconcile key-value data through failures and topology changes.

Distributed Data3.5K+ words · rationale · 2 diagrams
19 / 60

Payment Processing

Accept money with idempotency, an auditable ledger, and reconciliation across unreliable partners.

Transactions3.5K+ words · rationale · 2 diagrams
20 / 60

Shopping Cart & Checkout

Coordinate carts, inventory, pricing, orders, and payments without distributed transactions.

Transactions3.5K+ words · rationale · 2 diagrams
21 / 60

Collaborative Document Editor

Merge concurrent edits, broadcast presence, and persist recoverable document history.

Collaboration3.5K+ words · rationale · 2 diagrams
22 / 60

Cloud Object Storage

Store immutable objects with multipart upload, replication, versioning, and extreme durability.

Storage3.5K+ words · rationale · 2 diagrams
23 / 60

Distributed Message Queue

Provide ordered partitions, durable retention, consumer groups, and replayable offsets.

Distributed Data3.5K+ words · rationale · 2 diagrams
24 / 60

Feature Flag Service

Evaluate flags locally at low latency while configuration changes propagate safely.

Control Planes3.5K+ words · rationale · 2 diagrams
25 / 60

Recommendation System

Collect behavior, train representations, retrieve candidates, and rank them online.

ML Systems3.5K+ words · rationale · 2 diagrams
26 / 60

Fraud Detection Pipeline

Score risky events in real time while preserving explainability, audit, and model evolution.

ML Systems3.5K+ words · rationale · 2 diagrams
27 / 60

Configuration Management

Version, validate, watch, and roll out configuration without taking services down.

Control Planes3.5K+ words · rationale · 2 diagrams
28 / 60

Distributed Lock Service

Coordinate exclusive work using leases and fencing rather than trusting process ownership.

Coordination3.5K+ words · rationale · 2 diagrams
29 / 60

Multi-Tenant SaaS

Isolate tenants across identity, data, compute, quotas, and operations.

Platform3.5K+ words · rationale · 2 diagrams
30 / 60

Unique ID Generator

Generate sortable, globally unique identifiers without a central database bottleneck.

Foundations3.5K+ words · rationale · 2 diagrams
31 / 60

Parking Lot

Model entrances, spots, tickets, pricing, and concurrent allocation across floors.

Low-Level Design3.5K+ words · rationale · 2 diagrams
32 / 60

Elevator System

Schedule cars, serve hall and cabin requests, and preserve safety under concurrent events.

Low-Level Design3.5K+ words · rationale · 2 diagrams
33 / 60

Library Management

Manage catalog items, copies, members, holds, loans, fines, and notifications.

Low-Level Design3.5K+ words · rationale · 2 diagrams
34 / 60

Meeting Room Scheduler

Find rooms, reserve time ranges, invite attendees, and prevent overlapping bookings.

Low-Level Design3.5K+ words · rationale · 2 diagrams
35 / 60

ATM

Authenticate users, dispense cash, and coordinate account transactions with device state.

Low-Level Design3.5K+ words · rationale · 2 diagrams
36 / 60

Vending Machine

Accept payment, select products, dispense inventory, return change, and recover faults.

Low-Level Design3.5K+ words · rationale · 2 diagrams
37 / 60

Chess or Card Game

Represent legal moves, turns, game state, persistence, and optional multiplayer synchronization.

Low-Level Design3.5K+ words · rationale · 2 diagrams
38 / 60

Logging Framework

Design structured logging APIs, levels, appenders, formatting, buffering, and rotation.

Low-Level Design3.5K+ words · rationale · 2 diagrams
39 / 60

In-Memory LRU Cache

Provide O(1) get/put, eviction, capacity control, and optional expiration safely.

Low-Level Design3.5K+ words · rationale · 2 diagrams
40 / 60

Task Management System

Model projects, tasks, assignment, status workflows, dependencies, comments, and audit history.

Low-Level Design3.5K+ words · rationale · 2 diagrams
41 / 60

Web Search Engine

Crawl, index, rank, and serve the open web with fresh, trustworthy results.

Search3.5K+ words · rationale · 2 diagrams
42 / 60

Email Service

Accept, route, store, filter, and synchronize email across unreliable external domains.

Communication3.5K+ words · rationale · 2 diagrams
43 / 60

Live Streaming Platform

Ingest live video, transcode continuously, and deliver low-latency playback to massive audiences.

Media3.5K+ words · rationale · 2 diagrams
44 / 60

Advertisement Serving

Select eligible ads, pace budgets, and record billable events within a strict latency budget.

Advertising3.5K+ words · rationale · 2 diagrams
45 / 60

Stock Exchange

Accept orders, match deterministically, publish market data, and preserve an auditable ledger.

Transactions3.5K+ words · rationale · 2 diagrams
46 / 60

Food Delivery Marketplace

Coordinate menus, orders, restaurants, couriers, dispatch, and settlement in real time.

Geo & Realtime3.5K+ words · rationale · 2 diagrams
47 / 60

Hotel Reservation Platform

Search inventory and reserve room-night capacity without overselling across channels.

Transactions3.5K+ words · rationale · 2 diagrams
48 / 60

Digital Wallet

Hold balances, transfer value, enforce limits, and reconcile funding rails safely.

Transactions3.5K+ words · rationale · 2 diagrams
49 / 60

Expense Sharing

Model groups, split strategies, debts, settlements, and simplified balances without losing history.

Low-Level Design3.5K+ words · rationale · 2 diagrams
50 / 60

Online Auction

Run timed auctions, enforce bid ordering, extend deadlines, and settle winners consistently.

Transactions3.5K+ words · rationale · 2 diagrams
51 / 60

Maps and Navigation

Serve maps, geocode places, compute routes, and update travel times from live traffic.

Geo & Realtime3.5K+ words · rationale · 2 diagrams
52 / 60

Service Discovery

Register instances, publish health-aware endpoints, and provide stable discovery through partitions.

Control Planes3.5K+ words · rationale · 2 diagrams
53 / 60

Secrets Management

Store, authorize, rotate, lease, and audit secrets without exposing plaintext broadly.

Security3.5K+ words · rationale · 2 diagrams
54 / 60

CI/CD Platform

Trigger reproducible builds, schedule isolated runners, preserve artifacts, and deploy safely.

Developer Infrastructure3.5K+ words · rationale · 2 diagrams
55 / 60

Package Registry

Publish immutable packages, resolve versions, scan artifacts, and distribute them globally.

Developer Infrastructure3.5K+ words · rationale · 2 diagrams
56 / 60

Workflow Orchestrator

Run durable multi-step workflows that survive retries, timers, worker crashes, and long waits.

Infrastructure3.5K+ words · rationale · 2 diagrams
57 / 60

Distributed SQL Database

Provide SQL transactions over sharded, replicated data with online rebalancing.

Distributed Data3.5K+ words · rationale · 2 diagrams
58 / 60

Identity and Authentication

Authenticate humans and services, issue sessions/tokens, enforce MFA, and revoke compromised access.

Security3.5K+ words · rationale · 2 diagrams
59 / 60

Spreadsheet Engine

Model cells, formulas, dependencies, recalculation, cycles, and collaborative edits.

Low-Level Design3.5K+ words · rationale · 2 diagrams
60 / 60

File System

Design files, directories, inodes, block allocation, journaling, caching, and crash recovery.

Low-Level Design3.5K+ words · rationale · 2 diagrams
No system matches that filter. Try a broader concept.