Add GEPA Optimize_Anything skill eval and optimize Capability for Skill evaluation#166

Draft

auschoi96 wants to merge 24 commits intodatabricks-solutions:mainfrom

Commits on Feb 21, 2026

adding skill evals and skill optimization capabilities using GEPA optimize_anything
auschoi96
committed

Commits on Feb 24, 2026

addition and updates to evaluation powering GEPA. added capability to add examples from MLflow traces or manual
auschoi96
committed
readme updates
auschoi96
committed

Commits on Feb 26, 2026

Merge branch 'databricks-solutions:main' into main
auschoi96
authored

Commits on Mar 2, 2026

Merge branch 'databricks-solutions:main' into main
auschoi96
authored

Commits on Mar 3, 2026

add the parsing skill for parsing documents/custom rag

NatyraB
authored and
auschoi96
committed
feat: add lakehouse monitoring reference to Unity Catalog skill

Kevin Nguyen
authored and
auschoi96
committed
feat: updated lakehouse monitoring to use new naming data profiling

Kevin Nguyen
authored and
auschoi96
committed
updated client files with latest sdk syntax recommendations

dbderek
authored and
auschoi96
committed
install databricks parsing skill

NatyraB
authored and
auschoi96
committed
refactoring to use mlflow
auschoi96
committed
refactoring to use mlflow
auschoi96
committed
Merge remote-tracking branch 'upstream/main'
auschoi96
committed
added mlflow judges and ability to optimize skills and tools independently. This is because tools are used universally so we may not be able to optimize the two together
auschoi96
committed
lint fixes
auschoi96
committed
lint fixes
auschoi96
committed