Data Engineering Concepts

• Data vs Database vs Data Warehouse vs Data Lake: what each is actually for

You'll learn to: speak the vocabulary of data architecture correctly from day one.

• CSV, JSON, Avro, Parquet, ORC: structure, compression, and when to use each

You'll learn to: choose the right file format for a pipeline based on cost, performance, and downstream use case.

• Normalization vs Denormalization
• Fact Tables vs Dimension Tables
• Star Schema vs Snowflake Schema

You'll learn to: design warehouse schemas that are fast to query and easy to maintain. This is a top recurring interview topic.

• Types 0, 1, 2, 3, 4, and 6: how each handles historical data, and when to use it

You'll learn to: correctly implement change-tracking in dimension tables, a skill that separates junior from mid-level data engineers.

• Core characteristics of each approach, and the GCP services that support them

You'll learn to: decide when a pipeline needs to be real-time versus scheduled, and design accordingly.

• Tasks and dependencies
• Sequential, parallel, and conditional execution patterns

You'll learn to: design pipeline logic that is resilient, efficient, and easy to debug when something breaks.

• The layered pipeline design: Source → Landing → Raw → Curated → Target
• Fail-fast validation, error handling, and reusability across multiple pipelines

You'll learn to: architect a complete, production-grade ETL system from scratch. It is the single most valuable skill this program builds toward, and the one interviewers probe hardest.

• What level of SQL is actually needed (joins, window functions, partitioning/clustering)
• What level of Python is actually needed (data structures, reading/validating LLM-generated code)

You'll learn to: meet the real, practical bar companies expect: not academic mastery, but working fluency.