The Work Gym

Courses / Data and analytics tools

Databricks

Build reliable data pipelines in a workspace with SQL, PySpark, Delta Lake and Unity Catalog.

Start lesson 1 Practice interview Take the assessmentPractice the first unit of every course. No account needed.

57lessons

382checked exercises

14projects, cases and exams

Prepares you for Databricks Data Engineer Associate

Projects, case studies and mock exams

Project · IntermediateRetailer bronze to gold

Aster Supply's operations lead needs regional revenue that can be traced to individual orders.

6 steps · 18 min Project · IntermediateRide operations mart

Blue Heron Rides needs a route mart for its operations review.

6 steps · 18 min Project · IntermediateRecoverable bronze ingestion

Aster Supply's landing directory contains a change batch rather than a full order snapshot.

6 steps · 18 min Project · AdvancedCustomer history with MERGE

Aster Supply needs customer-region history for reports that use the region valid when an event happened.

6 steps · 18 min Project · AdvancedStreaming-style quality pipeline

Blue Heron Rides receives trip events from its dispatch system.

6 steps · 18 min Case study · AdvancedPartner access review

Aster Supply's East partner should see order ids and known amounts for completed East orders.

6 steps · 18 min Project · ExpertProduction release review

Blue Heron Rides has reviewed release rides-r42 for its daily route mart.

6 steps · 18 min Case study · ExpertRevenue reconciliation

Aster Supply's finance lead has found two different revenue totals for the same reporting period.

6 steps · 18 min Mock exam · ExpertData Engineer Associate mock A

Answer 45 original questions about engineering decisions across seven certification domains.

45 questions · 90 min Mock exam · ExpertData Engineer Associate mock B

Work through a second 45-question form with independent scenarios and explanations.

45 questions · 90 min

Practice Databricks in a browser workspace with notebooks, a SQL editor and a catalog explorer. Start with selections and missing values, then write PySpark transformations, join and rank records, and build Delta tables. Apply change feeds with MERGE, inspect committed versions and recover earlier data. Projects take original retail and transit data through bronze, silver and gold layers. You will also practice incremental files, declarative quality rules, task dependencies and precise data access. Easy, medium and hard PySpark interviews, a SQL interview and two original 45-question exams prepare you for engineering work and the Databricks Data Engineer Associate outline.

Projects that use Databricks

Work one company brief through several tools. Your Databricks stage reads the tables from the stage before.

Project · WayfriendFrom invites to meetupsDatabricks → dbt → SQL → Dagster · 3 hr 30 min Case study · WayfriendThe reminder questionDatabricks → SQL · 45 min Journey · WayfriendYour first 90 days as a data engineerDatabricks → dbt → SQL → Dagster · 8 hr

See all projects

Beginner

Workspace and SQL

Notebooks, compute, namespace and first queries.

3 lessons

1Workspace and attached computeRun notebook cells with the correct namespace and attached compute.4 exercises · 12 minFree

2Selections and Spark SQL typesChoose columns and compute typed results in Spark SQL.4 exercises · 12 minFree

Practice setWorkspace practiceReview cells, columns and counts against a separate outlet extract.Free

SQL for daily work

Filter, summarize and combine warehouse records.

4 lessons

4Filters, dates and missing valuesFilter orders with explicit status, date and missing-value rules.4 exercises · 12 minWith a plan

5Groups and business totalsSummarize order feeds without confusing orders, buyers and amounts.4 exercises · 12 minWith a plan

6SQL joins and row countsCombine orders with customers while keeping unmatched and repeated keys visible.4 exercises · 12 minWith a plan

Practice setSQL practiceCombine filters, aggregates and joins on the outlet extract.With a plan

PySpark notebook foundations

Read DataFrames and transform columns without collecting data.

4 lessons

8DataFrames and notebook outputRead tables and display a selected DataFrame in a Python cell.4 exercises · 12 minWith a plan

9Column expressions and conditionsAdd and rename columns with explicit NULL and text rules.4 exercises · 12 minWith a plan

10Missing values and duplicatesHandle missing values and duplicate keys without changing unrelated rows.4 exercises · 12 minWith a plan

Practice setPySpark foundations practiceApply selections, conditions and missing-value handling to the outlet feed.With a plan

Intermediate

Delta tables and updates

Create reliable Delta tables and apply change feeds.

6 lessons

12Delta table creationCreate Delta tables with explicit schemas and source-driven CTAS.4 exercises · 12 minWith a plan

13Changes with INSERT, UPDATE and DELETEChange the intended records while preserving the rest of a Delta table.4 exercises · 12 minWith a plan

14MERGE and source coverageApply updates, inserts and source-driven deletes with explicit match rules.4 exercises · 12 minWith a plan

15DeltaTable merges from PythonApply keyed changes with the DeltaTable merge builder and inspect the saved table.4 exercises · 12 minWith a plan

ProjectRetailer bronze to goldAster Supply's operations lead needs regional revenue that can be traced to individual orders.With a plan

Practice setDelta changes practiceReview writes and merges in an isolated working table.With a plan

PySpark transformations

Join, aggregate and rank DataFrames for business analysis.

5 lessons

18DataFrame joins and unionsCombine DataFrames at the intended grain and align union columns.4 exercises · 12 minWith a plan

19DataFrame aggregatesBuild regional measures with the correct grain and denominator.4 exercises · 12 minWith a plan

20Windows and deterministic rankingRank records and calculate running measures without collapsing rows.4 exercises · 12 minWith a plan

ProjectRide operations martBlue Heron Rides needs a route mart for its operations review.With a plan

Practice setDataFrame analysis practiceReview DataFrame joins, aggregates and windows on outlet data.With a plan

Files and incremental loading

Load files once and keep incremental ingestion recoverable.

5 lessons

23CSV, JSON and Parquet readsRead known files with the matching format and inspect their DataFrame output.4 exercises · 12 minWith a plan

24COPY INTO and file identityLoad governed files incrementally and understand what a rerun remembers.4 exercises · 12 minWith a plan

25Auto Loader and checkpointsUse cloudFiles with separate schema and checkpoint locations for finite incremental ingestion.4 exercises · 12 minWith a plan

ProjectRecoverable bronze ingestionAster Supply's landing directory contains a change batch rather than a full order snapshot.With a plan

Practice setIncremental ingestion practiceChoose recoverable file and connector patterns and check a loaded table.With a plan

Advanced

History and evolving tables

Recover Delta data and preserve dimension history.

5 lessons

28Delta history and recoveryInspect committed versions and restore the intended table snapshot.4 exercises · 12 minWith a plan

29Schema evolution and constraintsKeep schema changes deliberate and enforce row validity at write time.4 exercises · 12 minWith a plan

30Slowly changing dimensionsDistinguish current-value corrections from historical dimension tracking.4 exercises · 12 minWith a plan

ProjectCustomer history with MERGEAster Supply needs customer-region history for reports that use the region valid when an event happened.With a plan

Practice setHistory and schema practiceReview snapshot recovery, schema contracts and dimension history.With a plan

Declarative pipelines

Build streaming tables and materialized views with checked quality rules.

4 lessons

33Streaming tables and materialized viewsDeclare pipeline outputs and their dependencies instead of sequencing manual copies.4 exercises · 12 minWith a plan

34Expectations and rejected rowsEnforce quality predicates and distinguish dropped records from failed updates.4 exercises · 12 minWith a plan

ProjectStreaming-style quality pipelineBlue Heron Rides receives trip events from its dispatch system.With a plan

Practice setDeclarative pipeline practiceReview table declarations, quality policies and source dependencies.With a plan

Unity Catalog governance

Grant precise access and publish restricted reporting views.

4 lessons

37Catalog hierarchy and privilegesGrant the intended principal precise access and inspect the resulting grants.4 exercises · 12 minWith a plan

38Restricted views and governed sharingPublish approved columns and rows without exposing the unrestricted source.4 exercises · 12 minWith a plan

Case studyPartner access reviewAster Supply's East partner should see order ids and known amounts for completed East orders.With a plan

Practice setGovernance practiceReview least privilege, object lifecycle and restricted reporting access.With a plan

Expert

Jobs and deployment

Orchestrate production work and promote the same reviewed code.

4 lessons

41Task graphs, triggers and repair runsChoose task dependencies, triggers and recovery behavior for a production workflow.4 exercises · 12 minWith a plan

42Git folders and deployment bundlesPromote reviewed code with explicit environment configuration.4 exercises · 12 minWith a plan

ProjectProduction release reviewBlue Heron Rides has reviewed release rides-r42 for its daily route mart.With a plan

Practice setProduction operations practiceReview task graphs, recoverable runs and environment promotion.With a plan

Performance and complex data

Read Spark plans and transform nested data without unsupported speed claims.

6 lessons

45Partitions, layouts and Spark metricsInterpret physical-layout commands and performance evidence without invented benchmarks.4 exercises · 12 minWith a plan

46Arrays, JSON and Spark SQL dialectExtract JSON fields and reshape repeated values without losing their parent grain.4 exercises · 12 minWith a plan

47QUALIFY and window filtersFilter a completed window calculation while preserving its input rows.4 exercises · 12 minWith a plan

48PIVOT and category columnsTurn known status values into reporting columns at a deliberate grain.4 exercises · 12 minWith a plan

Case studyRevenue reconciliationAster Supply's finance lead has found two different revenue totals for the same reporting period.With a plan

Practice setPerformance and nested-data practiceReview complex-data grain and evidence-based performance choices.With a plan

Interviews and certification

Explain tradeoffs and practice two original timed exams.

7 lessons

Interview setPySpark interview: easyExplain basic DataFrame behavior and write short transformations that preserve missing data.With a plan

Interview setPySpark interview: mediumReason about join populations, aggregate denominators and window ordering.With a plan

Interview setPySpark interview: hardEvaluate recoverability, late changes and distributed execution tradeoffs.With a plan

Interview setDatabricks SQL interviewWrite audit queries and explain Delta behavior under missing values and repeated changes.With a plan

Mock examData Engineer Associate mock AAnswer 45 original questions about engineering decisions across seven certification domains.With a plan

Mock examData Engineer Associate mock BWork through a second 45-question form with independent scenarios and explanations.With a plan

Practice setExpert consolidation practiceResolve a fresh returns review with checked queries and operational decisions.With a plan