![rw-book-cover](https://substackcdn.com/image/fetch/$s_!ZzNS!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F97bd5d6e-5bdf-42d6-b36e-395f02fa05a3_3000x1706.png) ## Metadata - Author: [[Joe Reis]] - Full Title:: Where Data Engineering Is Heading in 2026 - 5+ Trends - Category:: #🗞️Articles - URL:: https://joereis.substack.com/p/where-data-engineering-is-heading - Read date:: [[2026-02-16]] ## Highlights > AI generates models on the fly. Who needs a semantic layer when LLMs can interpret messy schemas? At the rate models are progressing, nothing would surprise me at this point. ([View Highlight](https://read.readwise.io/read/01khjjr0nv0c62wn3y606z0b52)) > If we slice the data by role, **22%** of data engineers cite “lack of leadership direction” as a major issue. ([View Highlight](https://read.readwise.io/read/01khjjsgvn17qj7qzcekcz56mx))