Tailored System Design Study Roadmaps
Choose a structured learning track calibrated to your timeline — whether you have 1 week to cram before an interview or 3 months for Staff-level mastery.
1-Week High-Yield Interview Sprint
Tailored for engineers with interviews in 7 days. Focuses exclusively on high-impact architectural patterns: latency physics, L4/L7 load balancing, database indexing, sharding, consistent hashing, Raft, caching, Kafka, and top 4 interview designs.
Step-by-Step Curriculum Syllabus
Core Fundamentals & Latency Physics
Requirements scoping and hardware latency physics.
Scoping functional vs non-functional requirements (SLA, latency, throughput).
Memorize L1/RAM vs SSD vs Cross-datacenter RTT for sizing calculations.
Traffic Routing & Rate Limiting
Layer 4/7 proxies, SSL offloading, and traffic defense.
TCP vs HTTP routing, SSL termination, and reverse proxies.
Token Bucket & Sliding Window algorithms for API gateways.
Databases: Indexing, Sharding & Consistent Hashing
Scaling stateful storage horizontally with zero hot partitions.
B-Trees and clustered indexes turn O(N) full scans into O(log N) lookups.
Horizontal sharding strategies and mitigating hot partitions.
Consistent hashing rings with virtual nodes for minimal data migration.
Distributed Consensus & Caching
Consistency models, Raft leader election, and cache patterns.
CAP theorem: Consistency vs Availability during network partitions.
Raft consensus, heartbeats, randomized election timeouts, and majority quorum.
Cache-Aside vs Write-Through vs Write-Back strategies.
Asynchronous Messaging & Idempotency
Decoupled architectures with commit logs and safe retries.
Kafka commit logs, partitions, and consumer groups.
Idempotency keys for payment and ordering APIs.
Microservices Resilience & Fault Isolation
Preventing cascading failures across distributed microservices.
Circuit Breaker pattern states (Closed, Open, Half-Open).
Retry with exponential backoff and jitter.
Top Practice Problems & Interview Execution
End-to-end rehearsal of the two most asked system design questions.
Base62 encoding, Key Generation Service (ZooKeeper), and Redis cache tier.
Distributed token bucket rate limiter with Redis Lua scripts.