C#
System Design Interview Preparation — MAANG Complete Guide
40 system design problems with full High-Level Design documents, .NET Low-Level Design implementations, Mermaid architecture diagrams, practice questions, and interview tips.
How to Use This Repository
- Study a problem — Read the HLD doc, understand requirements, architecture, and trade-offs
- Review the code — Walk through the .NET implementation to internalize the algorithms
- Practice — Answer the 5 practice questions at the end of each problem without looking
- Mock interview — Pick a random problem, design it on a whiteboard in 35 minutes
Phase 1: Foundation Systems (Problems 1-5)
Core building blocks that appear as components inside every larger system.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 1 | API Rate Limiter | Part 1 / Part 2 / Part 3 | TokenBucket, SlidingWindow, Middleware, RedisStore | Token bucket, sliding window, Redis counters |
| 2 | Unique ID Generator | Part 1 / Part 2 / Part 3 | Snowflake, MachineIdProvider, LeaseRenewal | Snowflake IDs, clock skew, machine lease |
| 3 | Web Cache (CDN) | Part 1 / Part 2 / Part 3 | MultiLevelCache, Invalidation | L1/L2 cache, write-through, cache stampede |
| 4 | Distributed Counter | Part 1 / Part 2 | ShardedCounter, FlushWorker | Sharded writes, eventual read, flush worker |
| 5 | Load Balancer | Part 1 / Part 2 | LeastConnections, ConsistentHash, HealthCheck | L4/L7, consistent hashing, health checks |
Phase 2: Queue, Events & Reliability (Problems 6-10)
Asynchronous processing, reliable delivery, and distributed coordination.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 6 | Distributed Task Scheduler | Part 1 / Part 2 | TaskScheduler, Dispatcher | Delay queues, cron, idempotent execution |
| 7 | Webhook Delivery Service | Part 1 | DeliveryWorker | Exponential backoff, circuit breaker, DLQ |
| 8 | Distributed Messaging (Kafka) | Part 1 | — | Partitions, consumer groups, exactly-once |
| 9 | Distributed Locking | Part 1 | RedisLock | Redlock, fencing tokens, lease expiry |
| 10 | Batch Processing & Auditing | Part 1 | AuditConsumer, BatchVerifier | Event sourcing, CDC, idempotent consumer |
Phase 3: Social & Feed Systems (Problems 11-15)
Fan-out, ranking, and content delivery at social-media scale.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 11 | Twitter / X | Part 1 / Part 2 | Timeline, FanOut, Models | Fan-out on write vs read, celebrity hybrid |
| 12 | Part 1 | — | Image pipeline, explore ranking, stories | |
| 13 | Part 1 | — | Hot/Wilson ranking, nested comments, vote | |
| 14 | Facebook News Feed | Part 1 | — | ML ranking, EdgeRank, pull-on-demand |
| 15 | YouTube / Netflix | Part 1 | — | ABR streaming, encoding pipeline, rec engine |
Phase 4: Messaging & Real-Time (Problems 16-20)
WebSocket connections, presence, CRDTs, and real-time communication.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 16 | WhatsApp / Messenger | Part 1 / Part 2 | ChatService, Models | WebSocket gateway, message queue, E2EE, presence |
| 17 | Live Comments | Part 1 | — | Pub/sub, fan-out to WebSockets, throttling |
| 18 | Slack / Teams | Part 1 | — | Channels, threads, search, workspace isolation |
| 19 | Video Conferencing (Zoom) | Part 1 | — | SFU vs MCU, WebRTC, simulcast, SRTP |
| 20 | Google Docs (Collaboration) | Part 1 | — | OT vs CRDT, operation transform, conflict resolution |
Phase 5: E-Commerce & Payments (Problems 21-25)
Transactions, sagas, idempotency, and financial-grade consistency.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 21 | Amazon E-Commerce | Part 1 / Part 2 | OrderSaga, Models | Saga orchestration, compensating transactions |
| 22 | Payment Service (Stripe) | Part 1 | — | Idempotency keys, double-entry ledger, PCI |
| 23 | Digital Wallet | Part 1 | — | Balance consistency, optimistic locking, audit |
| 24 | Shopify Multi-Tenant | Part 1 | — | Multi-tenancy, tenant isolation, noisy neighbor |
| 25 | Food Delivery (DoorDash) | Part 1 | — | Real-time dispatch, geo-matching, ETA |
Phase 6: Data, Analytics & Observability (Problems 26-30)
Stream processing, probabilistic data structures, and lambda architecture.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 26 | Web Analytics (Google Analytics) | Part 1 | — | Lambda architecture, HyperLogLog, rollups |
| 27 | Metrics Monitoring (Prometheus) | Part 1 | — | TSDB, Gorilla compression, PromQL, alerting |
| 28 | Log Collection (ELK Stack) | Part 1 | — | Kafka buffer, ILM tiered storage, trace correlation |
| 29 | Ad Click Aggregation | Part 1 | — | Bloom + Redis dedup, watermarks, exactly-once |
| 30 | Top-K Request Analysis | Part 1 | CountMinSketch + TopKTracker | Count-Min Sketch, Space-Saving, min-heap |
Phase 7: Infrastructure & Distributed Systems (Problems 31-35)
The foundational distributed systems concepts that underpin everything.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 31 | Distributed KV Store (DynamoDB) | Part 1 | ConsistentHashRing, VectorClock | Consistent hashing, quorum, vector clocks, Merkle trees, gossip |
| 32 | Distributed File System (GFS) | Part 1 | — | Master/chunk architecture, 64MB chunks, write pipeline |
| 33 | Global CDN (CloudFront) | Part 1 | — | GeoDNS, Anycast, origin shield, cache invalidation |
| 34 | Distributed Tracing (Jaeger) | Part 1 | — | Trace/span/context, tail sampling, OpenTelemetry |
| 35 | Resource Allocation (K8s Scheduler) | Part 1 | PodScheduler | Filter-Score-Bind, bin packing vs spreading, preemption |
Phase 8: Boss Level (Problems 36-40)
Complex end-to-end systems that combine multiple patterns from previous phases.
| # | Problem | HLD Docs | .NET Code | Key Patterns |
|---|---|---|---|---|
| 36 | Uber/Lyft (Ride Sharing) | Part 1 | GeoSpatialIndex | GeoHash, driver matching, surge pricing, ride state machine |
| 37 | Ticketmaster (Event Ticketing) | Part 1 | — | Virtual waiting room, seat locking (FOR UPDATE), zero-oversell |
| 38 | Google Search | Part 1 | — | Crawler, inverted index, PageRank, fan-out query serving |
| 39 | Stock Trading Platform | Part 1 | OrderBook | Order book matching, price-time priority, event sourcing, double-entry ledger |
| 40 | Container Orchestration (K8s) | Part 1 | — | Control/data plane, reconciliation loops, etcd Watch, rolling updates |
Key Algorithms & Data Structures (Code Index)
Quick reference for the most important implementations.
| Algorithm | File | Used In |
|---|---|---|
| Token Bucket Rate Limiter | TokenBucketRateLimiter.cs | #1 Rate Limiter |
| Sliding Window Counter | SlidingWindowCounterRateLimiter.cs | #1 Rate Limiter |
| Snowflake ID Generator | SnowflakeIdGenerator.cs | #2 Unique ID |
| Multi-Level Cache (L1/L2) | MultiLevelCacheService.cs | #3 Web Cache |
| Sharded Counter | ShardedCounterService.cs | #4 Distributed Counter |
| Consistent Hash Load Balancer | ConsistentHashBalancer.cs | #5 Load Balancer |
| Least Connections Balancer | LeastConnectionsBalancer.cs | #5 Load Balancer |
| Redis Distributed Lock (Redlock) | RedisDistributedLock.cs | #9 Distributed Lock |
| Fan-Out Timeline Service | TimelineService.cs | #11 Twitter |
| Chat Service (WebSocket) | ChatService.cs | #16 Messenger |
| Saga Orchestrator | OrderSagaOrchestrator.cs | #21 E-Commerce |
| Count-Min Sketch + Top-K | CountMinSketch.cs | #30 Top-K |
| Consistent Hash Ring (vnodes) | ConsistentHashRing.cs | #31 KV Store |
| Vector Clock | VectorClock.cs | #31 KV Store |
| Pod Scheduler (Filter-Score-Bind) | SchedulerModels.cs | #35 Resource Allocation |
| GeoHash Spatial Index | GeoSpatialIndex.cs | #36 Uber |
| Order Book Matching Engine | OrderBook.cs | #39 Stock Trading |
Repository Stats
| Metric | Count |
|---|---|
| System Design Problems | 40 |
| HLD Documents | 52 |
| .NET Implementation Files | 53 |
| Practice Questions | 200 (5 per problem) |
| Architecture Diagrams | 80+ (Mermaid) |
Study Plan
| Week | Phase | Problems | Focus |
|---|---|---|---|
| 1 | Foundation | 1-5 | Core building blocks (rate limiter, cache, load balancer) |
| 2 | Queue & Events | 6-10 | Async processing, reliability, distributed coordination |
| 3 | Social & Feed | 11-15 | Fan-out strategies, ranking, content delivery |
| 4 | Messaging & RT | 16-20 | WebSockets, presence, CRDTs, real-time systems |
| 5 | E-Commerce | 21-25 | Transactions, sagas, payments, multi-tenancy |
| 6 | Analytics | 26-30 | Stream processing, probabilistic data structures |
| 7 | Infrastructure | 31-35 | Distributed systems fundamentals (the theory) |
| 8 | Boss Level | 36-40 | End-to-end complex systems (bring it all together) |
Cross-Cutting Concepts
These patterns appear across multiple problems. Master them once, apply everywhere.
| Concept | Problems Where It Appears |
|---|---|
| Consistent Hashing | #3 Cache, #5 Load Balancer, #31 KV Store, #33 CDN |
| Fan-Out | #11 Twitter, #14 Facebook, #17 Live Comments |
| Event Sourcing / CDC | #10 Auditing, #21 E-Commerce, #39 Stock Trading |
| Saga / Compensating Transactions | #21 E-Commerce, #22 Payments, #25 Food Delivery |
| Bloom Filter | #29 Ad Click, #30 Top-K, #38 Web Crawler |
| HyperLogLog | #26 Web Analytics |
| Count-Min Sketch | #29 Ad Click, #30 Top-K |
| GeoHash / Spatial Index | #25 Food Delivery, #36 Uber |
| Quorum (W + R > N) | #31 KV Store, #32 File System |
| Leader Election | #9 Distributed Lock, #32 File System, #40 Kubernetes |
| Circuit Breaker | #7 Webhooks, #22 Payments |
| Backpressure | #8 Kafka, #28 Log Collection, #29 Ad Click |
| Reconciliation Loop | #35 Scheduler, #40 Kubernetes |
| Double-Entry Ledger | #22 Payments, #23 Wallet, #39 Stock Trading |