A daily executive dashboard reads a fact table fed by three sources at three different natural caden

A medium Pipeline Design interview practice problem on DataDriven. Write and execute real pipeline design code with instant grading.

Domain

Pipeline Design

Difficulty

medium

Problem

A daily executive dashboard reads a fact table fed by three sources at three different natural cadences: mobile events from Kafka (continuous), Stripe payments (15-minute REST), and Salesforce CRM (daily). The current canvas forces all three through one mega-DAG running daily, making the Kafka events 16 hours stale. Apply the multi-cadence framing this section just taught and split into three upstream cadence-specific DAGs plus one downstream daily DAG. Add at least 2 distinct orchestration nodes (one for the streaming/micro-batch upstream cadences, one for the daily downstream join) and ensure three distinct freshness tiers are visible across the sources via slaFreshness labels (real-time/< 1min for Kafka, < 15min for Stripe, < 24h for Salesforce). The downstream daily DAG joins the three upstream raw assets at 5am via the lakehouse asset boundary.

Practice This Problem

Solve this Pipeline Design problem with real code execution. DataDriven runs your solution and grades it automatically.