Technical interview guide

Batch vs. Streaming Processing

Processing data in scheduled chunks versus continuously as it arrives — and the latency/complexity tradeoff.

Read: 28 min
Practice MCQs: 25
Interview QA: 25
Edition: v4
Editorial status: Reviewed
Relevant for: Data ArchitectData EngineerForward Deployed Engineer (FDE)

Scope: Apache Flink stable/current 2026-08-31; Apache Spark 4.2; Apache Beam, Kafka, Iceberg, OpenLineage current 2026-08-31.

Interview QA

Treat each question like a live interview question: answer out loud first (structure, assumptions, tradeoffs), then open the model answer to spot gaps and rehearse a tighter follow-up.

Curated: 2026-10-05 · Written: 2026-10-05 · Reviewed: 2026-10-03

QA-1

Choose batch, micro-batch, or streaming for a new data product.

QA-2

Design event-time windows and late-data behavior for a streaming metric.

QA-3

Build an end-to-end exactly-once business outcome.

QA-4

Handle backpressure in a high-volume streaming pipeline.

QA-5

Design partitioning and ordering for a keyed event stream.

QA-6

Design state management and checkpointing for a streaming application.

QA-7

Join two unbounded event streams safely.

QA-8

Build a streaming fraud-detection decision path.

QA-9

Build a hybrid streaming and batch reconciliation architecture.

QA-10

Migrate a batch pipeline to lower-latency streaming.

QA-11

Recover a streaming job after checkpoint corruption.

QA-12

Handle poison events without stalling a stream.

QA-13

Compare Lambda and Kappa architectures for data pipelines: when is Kappa preferred, and how do you handle historical data reprocessing?

QA-14

Walk me through how you decide between continuous streaming, micro-batch, and daily batch when designing an ingestion and transformation pipeline.

QA-15

Perform a safe stateful streaming application upgrade.

QA-16

Design streaming deduplication under retries and late arrivals.

QA-17

Control cost in a continuously running streaming platform.

QA-18

Design stream retention, replay, and consumer recovery.

QA-19

Design sessionization for user activity events.

QA-20

Design a streaming aggregation sink for analytical queries.

QA-21

Evolve a streaming event schema without breaking producers, consumers, or state.

QA-22

Design disaster recovery and regional failover for a streaming pipeline.

QA-23

Honor privacy deletion when immutable events feed batch and streaming systems.

QA-24

Design a capacity and correctness test for batch versus streaming processing.

QA-25