Best Data Engineering Newsletters to Follow (2026)

Data Engineering Weekly is the closest thing the field has to a default. Its about page describes a weekly aggregation with reviews instead of bare links, and it labels sponsored content as sponsored. The reviews are the product: reading 5 summaries tells you which 1 or 2 articles deserve your next half hour.

The Analytics Engineering Roundup is the pick when your work sits near the warehouse and the transformation layer. Tristan Handy curates it, and its own page describes the internet's most useful articles on analytics engineering and the tools around it. Its archive shows an irregular rhythm, with weeks of posts followed by gaps, so treat it as a feed you read when it arrives. If your stack runs on dbt, it pairs well with preparing for dbt interview questions.

Data Engineer Things is a community first and a newsletter second. Its page describes a newsletter on industry trends and events, with expert interviews and learning resources. Members write articles, and there is a Slack and a set of meetups. It suits you if you want to write or meet people, not only read.

Data News, published at blef.fr, numbers its posts by week and skips weeks, with at least one post announcing a break. Read its archive before you subscribe, so the rhythm isn't a surprise.

An author-led newsletter is a bet on a person. SeattleDataGuy's Newsletter, by Ben Rogojan, says it explores the technical side of the data world along with people and processes, and it is written for engineers and managers, plus consultants. Its sample topics include team structure, whether to use Airflow and getting promoted, which makes it the better pick for the questions a digest never answers: how teams are organised and how a career moves. Its archive shows a post every 2 to 3 weeks.

Practical Data Modeling suits the part of the job interviews probe hardest and day-to-day work neglects. Its page describes data modeling trends and practical advice for beginners and experienced professionals. It also offers a community, and early access to chapters of a book. Full access needs a paid subscription, so judge the free posts first. The matching practice is the data modeling interview questions page.

Simon Späti's newsletter and Start Data Engineering sit between the 2 types. Späti's page describes curated content on data engineering and productivity, plus digital note-taking, tied to his blog and book. Start Data Engineering is first a site of long, hands-on posts on pipelines, Spark, Iceberg and Airflow, with data quality coverage, interview prep and portfolio projects too, and a newsletter sign-up page. If you learn by building, its posts are worth more than most digests.

Decide by the question you want answered, not by the ranking. If the question is what changed this week, subscribe to a curated digest. If it is how do other teams do this, subscribe to an author whose work resembles yours. If it is what should I be learning next, read the digest for 4 weeks and note which topics you keep opening.

Check 3 things on the newsletter's own page before you subscribe: whether the archive is current, whether the writing you can read free is the kind you want, and whether sponsored content is labelled. A stale archive is a verdict by itself. Subscriber counts tell you about reach, not quality. They also change, so this page quotes none; read them at the source if they matter to you.

A starting pair that works for most engineers is 1 curated digest plus 1 author who works on your kind of problem. Warehouse and dbt work points to the Analytics Engineering Roundup, and modeling to Practical Data Modeling. For careers and team questions, pick SeattleDataGuy. Swap one every quarter if it stops teaching you anything.

The failure mode is an unread backlog that makes you feel behind. Fix it with a fixed slot: 30 minutes once a week, with a rule that anything older than 2 issues is archived unread. A digest that ships 5 reviewed items a week needs 30 minutes only if you read the reviews and open 1 or 2 links.

Keep a running note of what you learn, 1 line per item plus the system it concerns. A month of notes shows which tools and patterns recur, and that list is a better guide to what to practice for your own interviews than any ranking. Skip launches and funding news; keep posts that explain a failure or a migration, or weigh a tradeoff, because those are the stories interviewers ask you to tell about your own work.

A newsletter is one-way. When you have a question about your own pipeline, a community answers it. For interviews, a newsletter cannot tell you whether you can write a window function or draw a pipeline under a 45 minute clock, and the resources for data engineering interviews page lists the material that can.

Reading pays off when it turns into something you can explain. Take one idea from this week's issue, say a design for a late-arriving data fix, and sketch it in the data pipeline practice problems. Then check yourself against the data pipeline interview questions. If you are not sure what to learn next, the data engineer roadmap gives the order.