Back to glossary

Data Mesh

A decentralized data architecture paradigm where domain teams own and operate their data as products, with federated governance and self-serve infrastructure replacing centralized data teams.

Data mesh, introduced by Zhamak Dehghani, applies domain-driven design principles to data architecture. Instead of a central data team owning all data pipelines and the data warehouse, each domain team (payments, user engagement, content) owns their data end-to-end and publishes it as a product that other teams can discover and consume.

The four principles are domain ownership (teams own their data), data as a product (data is treated with the same rigor as customer-facing products), self-serve infrastructure (platform teams provide tools that domain teams use independently), and federated computational governance (global standards with local autonomy).

For AI teams in large organizations, data mesh addresses the bottleneck of centralized data teams. Instead of waiting weeks for a central team to build a pipeline for a new feature, the domain team that generates the data provides it as a well-documented, quality-assured data product. AI teams consume these data products as model inputs, with clear contracts around freshness, quality, and schema stability.

Related Terms