A modular curriculum with three entry paths, a neutral conceptual core, and a deep Databricks lakehouse track — engineered backward from the ten gaps that actually stall people. Data engineering is not "writing pipelines"; it is designing reliable data products you can defend.
One shared module catalog, three sequencings — each standalone, all ending on a defensible portfolio.
The full ~50-week course across four phases, ending on a governed capstone.
Start at Phase 1 →About five months, less if the diagnostic lets you skip more; it routes you to the analyst or SWE variant.
See the Bridge path →A self-diagnostic against the ten gaps, plus eight pre-built bundles.
See the Sharpen path →38 units across 12 domains. The neutral core is taught first; the Databricks/lakehouse track is where it gets implemented deeply. Fit columns show how each path treats a module.
| Module | Weeks | Launch | Bridge·A | Bridge·S |
|---|---|---|---|---|
| D0 Orientation & tooling | ||||
| M00 · Dev environment, CLI & Git | 1 | core | core | skip |
| M01 · The DE mental model & data lifecycle | 1 | core | core | core |
| D1 Programming — Python | ||||
| M02 · Python fundamentals for DE | 3 | core | core | skip |
| M03 · Python for data — files, APIs, pandas | 2 | core | core | comp |
| M04 · Software engineering hygiene | 1 | core | core | skip |
| D2 SQL & relational | ||||
| M05 · SQL fundamentals | 2 | core | skip | core |
| M06 · Advanced SQL — windows, CTEs, plans | 2 | core | comp | core |
| M07 · Relational theory & normalization | 1 | core | comp | core |
| D3 Data modeling & warehousing · closes G2 | ||||
| M08 · Dimensional modeling — grain, facts, dims | 2 | core | core | core |
| M09 · SCD & history tracking G3 | 1 | core | core | core |
| M10 · Warehouse/lakehouse & the gold layer | 1 | core | core | core |
| M11 · Reference data, crosswalks, MDM & entity resolution G2 | 2 | core | core | core |
| M12 · Data Vault & advanced patterns | 1 | elec | elec | elec |
| D4 Pipelines & ingestion · closes G3 | ||||
| M13 · ETL vs ELT, batch vs incremental | 1 | core | core | core |
| M14 · Ingestion patterns & source systems | 1 | core | core | core |
| M15 · Medallion architecture | 1 | core | core | core |
| M16 · CDC, watermarks, snapshots & backfills G3 | 2 | core | core | core |
| D5 Distributed compute — Spark · closes G7 | ||||
| M17 · Spark fundamentals & the PySpark DataFrame API | 2 | core | core | core |
| M18 · Spark internals — partitions, shuffles, skew | 1 | core | core | core |
| M19 · Performance tuning & cost | 1 | core | core | core |
| D6 Lakehouse platform — Databricks · closes G5 | ||||
| M20 · Delta Lake & table formats | 1 | core | core | core |
| M21 · Databricks & Unity Catalog fundamentals | 1 | core | core | core |
| M22 · Declarative pipelines (Lakeflow/DLT) & Jobs | 2 | core | core | core |
| M23 · Structured Streaming semantics G5 | 2 | core | core | core |
| D7 Data quality, testing & observability · closes G4 | ||||
| M24 · Data quality — expectations, quarantine, reconciliation G4 | 2 | core | core | core |
| M25 · Testing taxonomy for data — pytest | 1 | core | core | core |
| M26 · Observability, monitoring, SLAs & lineage | 1 | core | core | core |
| D8 DevOps & engineering practice · closes G6 | ||||
| M27 · Git deep — branching, PRs, review | 1 | core | core | skip |
| M28 · CI/CD for data — Actions, env promotion | 1 | core | core | comp |
| M29 · Databricks Asset Bundles (DABs) & IaC G6 | 1 | core | core | core |
| D9 Governance, security & cloud · closes G8 | ||||
| M30 · Governance & security — UC RBAC, masking, PII/PHI G8 | 1 | core | core | core |
| M31 · Cloud foundations — Azure or AWS track | 1 | core | core | comp |
| D10 Architecture & AI-assisted DE · closes G9 | ||||
| M32 · Architecture decision-making G9 | 1 | core | core | core |
| M33 · AI-assisted data engineering | 1 | elec | elec | elec |
| D11 Career & proof · closes G1 · G10 | ||||
| M34 · Communicating your engineering work G10 | ◇ | core | core | core |
| M35 · Portfolio engineering | ◇ | core | core | core |
| M36 · Interview prep — SQL, design, behavioral G1 | 2 | core | core | core |
| M37 · Capstone — end-to-end governed data product | 3–4 | core | core | core |
Bronze → silver → gold → launch. Each phase ends in a portfolio artifact; the path ends in a governed data product.
Mined from hundreds of real coaching calls — where people with certs and theory still stall. Every module is tagged to the gap it closes.