Senior answers.Staff judgment.Principal scope.
Every level can draw the boxes. What separates Senior, Staff and Principal answers is what you notice before you draw them and what you own after. Study the answer one level above where you are now.
“What are we protecting: infrastructure from abuse, paid quotas, or tenants from each other? They fail in opposite directions. For abuse I'd go local-first and fail-open, because a Redis hop on every request is a 5–10ms tax and a kill switch.”
Names the intent, the latency trap and who owns the failure mode. Clears the Staff bar.
Suggested order
Each step builds on the one before it. Skip ahead if the earlier material is already second nature.
- Foundations
Latency numbers, consistency, hashing and estimation. The vocabulary the rest of the library assumes.
- Patterns
Contention, scaling reads and writes, degraded mode. Reusable moves for any prompt.
- Core Ten
Rate limiting, caching, queues, locks, IDs, idempotency, multi-region and more. Practice the judgment loop on these first.
- Product case studies
Uber, WhatsApp, Ticketmaster and more, each with a Principal lens on cost, standards and multi-year evolution.
Core Ten
All case studiesTen problems that show up in almost every Staff loop, either as the whole prompt or as the hard part of a bigger one. Master these first.
- Rate LimitingDesign a Distributed Rate Limiter
- Distributed CachingDesign a Distributed Cache
- Message QueuesDesign a Distributed Message Queue
- Circuit BreakersDesign for Fault Tolerance
- Service DiscoveryDesign Service Discovery
- ID GenerationDesign a Distributed ID Generator
- Idempotency & Exactly-OnceDesign Idempotent APIs
- Distributed Lock ServiceDesign a Distributed Lock Service
- Multi-Region Active-ActiveDesign for Multi-Region Active-Active
- Ad Click AggregatorDesign a Real-Time Ad Click Aggregator
Latency-Sensitive
Every millisecond is someone's budget. Who spends it?
Data-Heavy
Volume changes the shape of every decision.
Correctness-Critical
Where a lost update is a lawsuit, not a blip.
Product Case Studies
The classic "Design X" prompts, taught through the Staff and Principal lens.
- Distributed Job SchedulerDesign a Distributed Job Scheduler73 min · 6 diagrams
- LeaderboardDesign a Leaderboard63 min · 6 diagrams
- Notification SystemDesign a Notification System60 min · 6 diagrams
- Replicated Data StoreDesign a Replicated Data Store73 min · 8 diagrams
- Stock ExchangeDesign a Stock Exchange66 min · 7 diagrams
- URL ShortenerDesign a URL Shortener66 min · 7 diagrams
- Web CrawlerDesign a Web Crawler73 min · 8 diagrams
- Blob StorageDesign Blob Storage (S3)71 min · 7 diagrams
- Metrics & MonitoringDesign Datadog66 min · 7 diagrams
- File SyncDesign Dropbox78 min · 6 diagrams
- Auto-Scaling & CapacityDesign for Auto-Scaling71 min · 8 diagrams
- Database SelectionDesign for Database Selection73 min · 8 diagrams
- Collaborative EditingDesign Google Docs70 min · 7 diagrams
- Maps & GeospatialDesign Google Maps62 min · 7 diagrams
- News FeedDesign Instagram Feed56 min · 6 diagrams
- Code ExecutionDesign LeetCode70 min · 8 diagrams
- Real-Time UpdatesDesign Real-Time Updates (WebSockets)65 min · 7 diagrams
- Flash Sales & TicketingDesign Ticketmaster61 min · 7 diagrams
- Proximity MatchingDesign Tinder75 min · 7 diagrams
- Ride Hailing & DeliveryDesign Uber66 min · 7 diagrams
- Chat MessagingDesign WhatsApp64 min · 6 diagrams
- Search TypeaheadDesign Search Autocomplete74 min · 8 diagrams
Foundations
Open sectionThe mental models Staff answers are built on. Fast to read, slow to master.
- Network Latency & Protocols34 min · 5 diagrams
- Caching Fundamentals38 min · 6 diagrams
- Consistent Hashing31 min · 6 diagrams
- Consistency Models & Partition Behavior31 min · 5 diagrams
- Back-of-Envelope Estimation29 min · 4 diagrams
- API Design Patterns31 min · 4 diagrams
- Database Indexing31 min · 4 diagrams
- Data Modeling & Schema Design31 min · 4 diagrams
- Sharding & Partitioning31 min · 4 diagrams
Frameworks & Patterns
Open sectionCross-cutting patterns you reuse in every design. Learn once, apply everywhere.
- Scaling Reads35 min · 5 diagrams
- Scaling Writes37 min · 5 diagrams
- Dealing with Contention39 min · 5 diagrams
- Real-time Updates37 min · 5 diagrams
- Data Pipeline Patterns37 min · 5 diagrams
- Degraded Mode Framework36 min · 5 diagrams
- Distributed Coordination35 min · 5 diagrams
- Build vs Buy Framework38 min · 6 diagrams
- Handling Large Blobs34 min · 6 diagrams
- Managing Long-Running Processes35 min · 5 diagrams
Key Technologies
Open sectionWhat each tool is actually good at, how it fails, and when to say no.
- Redis42 min · 6 diagrams
- PostgreSQL41 min · 5 diagrams
- Elasticsearch38 min · 6 diagrams
- Cassandra43 min · 5 diagrams
- DynamoDB40 min · 5 diagrams
- Time Series DBs39 min · 5 diagrams
- Apache Kafka38 min · 7 diagrams
- Flink & Stream Processing36 min · 6 diagrams
- API Gateways44 min · 6 diagrams
- ZooKeeper & etcd37 min · 6 diagrams