Data Engineering Projects (Real Pipelines, Real Portfolio)

The best data engineering projects look like real pipeline work. These are real data systems you fix in a live cloud workspace - a non-idempotent load, a slow report, a broken dbt model - each one a portfolio piece.

Why these beat a Kaggle notebook

A notebook shows analysis; data engineering is about reliable pipelines and correct data at scale. These practice exactly that, the part that gets you hired as an engineer.

18 projects to start with

Build a Kimball Star SchemaData · Senior Build an Exactly-Once Kafka PipelineData · Senior Switch Batch ETL to Streaming on AWS KinesisData · Senior Build a Data Quality Monitoring APIData · Mid Schedule a Pipeline as an Airflow DAGData · Mid Build a Staging-to-Marts dbt ProjectData · Mid Implement SCD2 With dbt SnapshotsData · Mid Migrate a pandas Pipeline to PolarsData · Mid Implement the Outbox Pattern (At-Least-Once)Data · Mid Make an ETL Pipeline IdempotentData · Junior Speed Up a Slow SQL Report QueryData · Junior Make a CSV Importer Resilient to Bad DataData · Junior Convert Timestamps to Local Time (Timezones)Data · Junior Implement Incremental Loading in an ETLData · Junior Validate DataFrames With panderaData · Junior Convert CSV to Parquet for a LakehouseData · Junior Query a Parquet Lake With DuckDBData · Junior Rank Rows Per Group With SQL Window FunctionsData · Junior

Pick by the skill to show

Reliability? The idempotency and pipeline ones. SQL depth? The window-function and tuning ones. Each finished fix is evidence on your portfolio.

Build your portfolio free. You are in a real cloud workspace in 30 seconds - fix real systems, show real work.

Start free →