Post

Jacob Matson
@matsonj.com

Extremely proud of this blog I wrote about Dual Execution on @motherduck.com - it brings the magic of insanely scalable cloud for prod and fast iteration on local to dbt in a seamless way. Very happy to share this with the world. motherduck.com/blog/dual-ex...

motherduck.comLocal dev and cloud prod for faster dbt development - MotherDuck BlogSpark the Joy of beautiful local development workflows with MotherDuck & dbt | Reading time: 7 min read

3 reposts25 likes

StretchArmstrong
@stretcharm1.bsky.social

Nice. I'm also doing this with dbt but with duckdb and databricks. Some minor macros to help with some syntax diffs. However the dbt sdf news means that might be a thing of the past. I've also got most of it running in dagster

StretchArmstrong
@stretcharm1.bsky.social

Only irritations are I like to use dbeaver to check my data and I have to keep disconnecting to run dbt. On occasion even need to quit dbeaver to release the file lock. I've also see dbt get locked out of the db by itself which is very odd as everything is single thread.

StretchArmstrong
@stretcharm1.bsky.social

And lastly if I host my parquet files in azure its way slower to run the pipe than if the files are local. All that said its way faster and cheaper to do my dev than having to start and stop Databricks. Loving the duck