Re-engineer something you own
Rebuild something already in production and prove the new form behaves like the old one.
14 Recipes
Cookbook
120 Recipes. Each one is a requirement-sized piece of data engineering work — a single outcome you can name, adopt, and review against acceptance conditions written down before the work starts. Recipes compose; the Cookbook is what they compose from.
Written for data engineers.
Recipe definitions from vibedata-cookbooks.
Browse by job
Every Recipe sits in exactly one job. The job is the kind of work, not the subject of it.
Build something that is not there yet
Create a pipeline, model, or data product that does not exist yet, with its proof attached.
99 Recipes
Prove a change is safe before you ship
Establish whether an existing change is safe or equivalent without changing it further.
1 Recipe
Explain a number after you shipped
Work out why an observed number, result, or cost came out the way it did.
1 Recipe
Work across more than one platform
Deliver an outcome whose defining difficulty is that it spans more than one platform.
5 Recipes
Collections
A Collection is a view over the same Recipes, not a copy of them. The same Recipe turns up in several, and it is the same Recipe in each.
By domain
The domain whose data you are working on today. You are still the data engineer in every one of these.
Larger outcomes
Recognisable pieces of work that take more than one Recipe, assembled rather than merged.
By problem area
The kind of problem the Recipes solve.
By industry
Vertical Recipes together with the horizontal ones that vertical reaches for most.
Browse by platform
Where the work can actually run. Each Recipe names its exact execution targets.
DuckDB
Recipes that can execute against local DuckDB.
112 Recipes
Microsoft Fabric
Recipes that can execute against Microsoft Fabric Lakehouse, Fabric Warehouse, or both.
97 Recipes
MotherDuck
Recipes that can execute against MotherDuck.
95 Recipes