The shift from monolithic to decentralized data workflows changes how teams build, connect and scale pipelines.

In this episode, we feature Oscar Ligthart, Lead Data Engineer, and Rodrigo Loredo, Lead Analytics Engineer, both at Vinted, as we unpack their YAML-driven abstraction that generates Airflow DAGs and standardizes cross-team orchestration.

Key Takeaways:

00:00 Introduction.

05:28 Challenges of decentralization.

06:45 YAML-based generator standardizes pipelines and dependencies.

12:28 Declarative assets and sensors align cross-DAG dependencies.

17:29 Task-level callbacks enable auto-recovery and clear ownership.

21:39 Standardized building blocks simplify upgrades and maintenance.

24:52 Platform focus frees domain work.

26:49 Container-only standardization prevents sprawl.

Resources Mentioned:

Oscar Ligthart

https://www.linkedin.com/in/oscar-ligthart/

Rodrigo Loredo

https://www.linkedin.com/in/rodrigo-loredo-410a16134/

Vinted | LinkedIn

https://www.linkedin.com/company/vinted/

Vinted | Website

https://www.vinted.com/?srsltid=AfmBOor87MGR_eLOauCO93V9A-aLDaAhGYx9cnu_oN8s1SAXMlCRuhW7

Apache Airflow

https://airflow.apache.org/

Kubernetes

https://kubernetes.io/

dbt

https://www.getdbt.com/

Google Cloud Vertex AI

https://cloud.google.com/vertex-ai

Airflow Datasets & Assets (concepts)

https://www.astronomer.io/docs/learn/airflow-datasets

Airflow Summit

https://airflowsummit.org/

Thanks for listening to “The Data Flowcast: Mastering Apache Airflow® for Data Engineering and AI.” If you enjoyed this episode, please leave a 5-star review to help get the word out about the show. And be sure to subscribe so you never miss any of the insightful conversations.

#AI #Automation #Airflow #MachineLearning