System Design Interview Rubric & Swim Lane Checklist
Complete architectural checklist, probing questions, formulas, and library tradeoffs across all 16 layers.
Client & Mobile Ingress
Native mobile apps (iOS/Android) and Web SPAs generating concurrent user traffic.
CDN & DNS Anycast Edge
Global Anycast DNS and edge Points of Presence absorbing static assets and GET cacheable APIs.
API Gateway & Edge Security
Centralized entry point managing JWT auth, rate limiting token buckets, SSL termination, and gRPC routing.
Load Balancer & Network Tier
Layer 4 (NLB) & Layer 7 (ALB) distributing ingress traffic across pods with health checks.
Backend Microservices & K8s
Stateless Kubernetes microservice pods communicating internally over low-latency binary gRPC.
Kafka Event Bus & Pub/Sub
Distributed event log decoupling synchronous order placement from asynchronous fulfillment.
Stream Processing & CDC (Debezium)
Real-time event stream transformations, change data capture, and analytical indexing.
Relational Database (PostgreSQL / Aurora)
ACID relational database with single primary for serializable writes and read replicas for hot reads.
NoSQL & Distributed Storage
Distributed wide-column and document stores for high-throughput write-heavy workloads.
In-Memory Cache (Redis Cluster)
Sub-millisecond distributed cache for session state, catalog SKU lookups, and inventory counters.
Message Queues (RabbitMQ / SQS)
Point-to-point task queues for background job distribution, emails, and async worker pools.
Dead Letter Queue (DLQ) & Retries
4-tier exponential backoff retry topics and persistent DLQ with alert webhooks for failed transactions.
Downstream Systems & 3rd Parties
External payment gateways, fraud APIs, logistics webhooks, and timeout budgets.
Observability, Logs & Tracing
Prometheus metrics, OpenTelemetry distributed tracing, OpenSearch structured logging, and SLOs.
Security, Zero Trust & Compliance
mTLS service mesh, KMS envelope encryption, IAM least privilege, PII data sanitization.
Scalability & Disaster Recovery
Multi-AZ active-active failover, RTO/RPO targets, database sharding, and cost optimization.