Search DistillSys

Find a concept

Type at least two characters to search lessons, designs, papers, and interview prep.

Applied reasoning · Scenario Lab

Make the call
under pressure.

Work through realistic incidents with incomplete evidence and competing constraints. Each exercise asks you to diagnose the failure, stabilize the system, recover safely, and prevent recurrence.

8deep scenarios4reasoning phases10rubric points each
Scenario library

Choose an incident

Showing 8 scenarios

01
Senior25 minNot started

Reliability · Incident commander

Checkout is melting under a retry storm

A slow payment dependency turns ordinary retries into a cascading failure across checkout.

Failure and retry patternsCircuit breakersLoad shedding
Enter scenario →
02
Senior25 minNot started

Consistency · Service owner

Users see stale balances after regional failover

Failover restores availability, but asynchronous replicas expose older account state.

Geo-replicationRPO and RTOConsistency models
Enter scenario →
03
Intermediate20 minNot started

Consensus · On-call engineer

A Raft cluster keeps changing leaders

Uneven latency and pauses create election churn without an obvious node failure.

RaftLeader electionQuorums
Enter scenario →
04
Intermediate20 minNot started

Partitioning · Platform engineer

One celebrity account overwhelms a shard

A technically balanced keyspace collapses when one key becomes globally hot.

Hot partitionsConsistent hashingRebalancing
Enter scenario →
05
Senior25 minNot started

Messaging · Application architect

Orders are fulfilled twice after a consumer restart

At-least-once delivery meets a non-idempotent warehouse side effect.

Delivery semanticsIdempotent workflowsDistributed logs
Enter scenario →
06
Staff+35 minNot started

Architecture · Principal engineer

Two regions sell the last unit

Active-active inventory remains available through a partition and violates uniqueness.

Active-activeConflict resolutionPACELC
Enter scenario →
07
Staff+30 minNot started

Coordination · Distributed systems engineer

A clock jump creates two lease holders

Time-based ownership outlives its safety assumptions after virtualization pauses and clock correction.

Time and orderingLeader electionIdempotent workflows
Enter scenario →
08
Senior25 minNot started

Messaging · Data platform owner

A slow sink stalls the event pipeline

Consumer lag, oversized batches, and unbounded buffering turn degradation into data loss risk.

BackpressureConsumer groupsLoad shedding
Enter scenario →