Sharpen is for working data engineers who "missed things along the way." Start with the self-diagnostic against the ten recurring gaps, then take individual modules or a pre-built bundle. Every unit stands alone and links straight to its lesson — no need to have done the earlier modules. The weekly live review is optional: bring anything you want to pressure-test.
Each gap maps to the modules that close it. Pick the ones you score low on; the module codes link to their lessons.
Eight curated gap-clusters. Each chip opens the real lesson week in the Launch phase pages.
Kill the full-refresh habit: watermarks, snapshots, backfills, idempotency.
Build the scale/shuffle instinct and know when tuning is actually worth it.
Grain, history, semantic layer, MDM, and Data Vault when it earns its place.
Expectations, quarantine, reconciliation, and the full test taxonomy.
Git → DABs → CI/CD with no manual drift and clean env promotion.
Unity Catalog design, RBAC, masking/hashing, PII/PHI, service identities.
Triggers, state, keys, idempotency — and why streaming is not real-time.
Choose and defend approaches; narrate the work to any audience.