Handle schema evolution in production.

Section 1 — The Context (The 'Why')
Schema evolution in production requires backward compatibility and coordinated rollout. The primary challenge is adding columns or deprecating without breaking consumers. Breaking changes cause production outages when consumers cannot parse new formats.

Section 2 — The Diagram

[New Schema] --> [Compat Check] --> [Deploy Producer]
      |                    |
      v                    v
[Registry]          [Consumer Update]

Section 3 — Component Logic
Additive changes (new optional columns) are safe and do not require consumer coordination. Breaking changes (removing columns, changing types) require a migration plan: schema registry validates compatibility, dual-write during transition, and deprecation with grace period. Avro and Protobuf support evolution; JSON Schema requires careful versioning. Idempotent consumers can handle multiple schema versions during rollouts. In production, monitor consumer lag, checkpoint success rate, and sink write latency as primary SLOs. Partitioning strategies should align with query patterns; bucketing within partitions mitigates join skew. TTL policies on raw and intermediate data control storage cost while preserving replay capability for debugging and backfill. Data skew mitigation via salting or secondary hashing prevents single partitions from becoming bottlenecks. Exactly-once semantics require transactional commits at the sink; at-least-once delivery demands idempotent write logic to avoid duplicates. Fan-out patterns allow one source topic to feed multiple downstream consumers without re-ingestion. Backpressure handling ensures that slow processors do not cause unbounded buffer growth; Kafka consumer lag is a key metric. Schema evolution should follow additive-only rules where possible to avoid breaking consumer compatibility. The CAP trade-off should be documented per component: analytics typically favors AP, while financial reconciliation requires CP. Blast radius from component failure is bounded by replication and checkpointing; design for graceful degradation during partial outages. Cost optimization: use Spot instances for batch workloads and tier cold data to lower storage classes. Dead-letter queues preserve failed records for replay rather than dropping them.

Section 4 — The Trade-offs (The 'Senior' part)