Engineering / advanced
Engineer systems that withstand failure
Design admission, retries, fault testing, recovery, and degraded operation around measurable service behavior.
Your route through engineering
Follow the path or begin with a question. Every lesson is open.
Your place, at your pace. Progress is optional and saved only in this browser when available.
Read / explore / reason
Inside this module
Module 01 advanced
Engineer systems that withstand failure
Design admission, retries, fault testing, recovery, and degraded operation around measurable service behavior.
- 01
Failure models and dependency graphs
Model required dependencies, shared failure domains, and correlated outages before interpreting a replica count as resilience.
31 min - 02
Tail latency and hedged requests
Analyze fan-out latency, conditional hedging, and the resource amplification hidden behind faster individual responses.
32 min - 03
Load shedding and admission
Design bounded admission around resource cost, deadlines, and tenant fairness instead of allowing overload to consume the service.
32 min - 04
Deadlines, retries, and idempotency
Compose bounded attempts with propagated deadlines and a durable idempotency boundary that preserves operation semantics.
34 min - 05
Chaos engineering and fault injection
Design bounded failure experiments with explicit hypotheses, independent stop controls, and recovery evidence.
33 min - 06
Multi-region failover
Plan regional traffic movement together with write authority, replication state, spare capacity, and a controlled failback.
35 min - 07
Disaster recovery and restoration
Treat backups as restoration inputs and validate recovery objectives against data consistency, access, throughput, and application checks.
34 min - 08
Dependency budgets and error budgets
Allocate end-to-end failure allowance across dependency paths and interpret burn rates without double-counting incidents or hiding cohorts.
33 min - 09
Production readiness and runbooks
Build evidence-based launch gates and executable operational guidance that remain useful when the normal system is failing.
32 min - 10
Safe degradation and fallbacks
Design cheaper useful responses with explicit freshness, authorization, quality, and recovery boundaries instead of hiding errors behind success codes.
34 min