System Design & Distributed Systems Bootcamp Certificate

Checkpoint 1: Write a system brief with functional/nonfunctional requirements, scale assumptions, latency/availability targets and failure budget.
Checkpoint 2: Prototype an RPC-based service pair; add timeouts, retries and idempotency, then demonstrate at least one concurrency/failure bug and fix.
Checkpoint 3: Design replicated state using primary/backup or quorum-style reasoning; document consistency, recovery and stale-read tradeoffs.
Checkpoint 4: Implement or simulate a small Raft-style consensus workflow and show leader election, log agreement and behavior under node failure.
Checkpoint 5: Add coordination/transactions for one cross-service workflow; reason about locking, atomicity, partial failure and exactly-once illusions.
Checkpoint 6: Produce capacity estimates and an architecture with load balancing, partitioning, caching/queues and observability; run a failure drill.

Day 1: Choose a realistic service and define requests/sec, data volume, latency SLO, availability, consistency needs and top failure modes.
Day 2: Build a small RPC/service prototype; instrument calls and add timeouts, retries, idempotency keys and backpressure assumptions.
Day 3: Add replicated state and a failure simulation; explain consistency, failover, recovery and what clients observe during disruption.
Day 4: Implement or simulate leader election/log replication; inject node/network faults and capture consensus/recovery evidence.
Day 5: Design one multi-service transaction/coordination flow and compare at least two approaches for atomicity and failure handling.
Day 6: Complete capacity model and production architecture; add caching/queue/partitioning where justified plus traces, metrics and SLOs.
Day 7: Publish repo/simulation, architecture diagrams, ADRs, capacity math, failure report, tradeoff memo, demo video and operations runbook.