Somewhat Analytics - 5th Edition

NEWSLETTER UPDATE: After two years of hosting my newsletter on LinkedIn, I'm migrating over to Substack! If you'd like to continue reading the Somewhat Analytics Newsletter, please subscribe the the Substack ?? here ??

Welcome to another edition of Somewhat Analytics - The Newsletter!

Simply put, this monthly newsletter is a brain dump of resources that I find interesting in the realm of data analytics and analytics engineering. These resources come in various formats such as blogs, tutorials, podcasts, social posts, events, my personal thoughts, and more. Why set any limits?

Let's dive in!??

(Blog) Evaluating my fitness by analyzing my Fitbit data archive by Olivia Tanuwidjaja (Analytics, Google)

I love when people create data projects outside of work. Personal data projects are indicators that individuals genuinely enjoy their craft. Plus, they often delve into interesting and less mundane topics, such as fitness.

Olivia recently analyzed her personal Fitbit data and discovered some compelling insights, including:

??♀? Which physical activities burn the most calories

?? How sleep affects her physical health

This is just one of Olivia's various "analytics geek"blogs. You can explore more of her writings here.

If you're interested in more personal Fitbit analytics projects, check out Zach Wilson’s project repository from last year. ??

(LinkedIn Post) 3 queries I run on a data table when adding dbt generic tests

Madison Schott is an experienced analytics engineer and the author of one of my favorite newsletters, Learn Analytics Engineering.

In this post, Madison shares three queries she runs before creating generic dbt tests. ??

Turn SQL templates into data APIs?!

I recently discovered VulcanSQL. It's a free, open-source data API framework that helps data teams generate API endpoints for their data consumers.

Creating these API endpoints doesn't require backend expertise. If you're comfortable with Jinja templates, you can make it happen.

I haven't tried the product yet, but it has definitely piqued my interest. If you've implemented it, I'd love to hear about your experience. Shoot me a message!

(Tutorial) How Snowflake EXCLUDE and ILIKE protects you from bad SQL. ILIKE it like that by Uli Bethke (Co-founder, Sonra)

Like other programming languages, SQL has its share of "anti-patterns," which are commonly used but ineffective solutions to problems.

Uli's article identifies various SQL anti-patterns (e.g., SELECT *) and explains how to replace these anti-patterns with high-quality, efficient alternatives (e.g., SELECT * EXCLUDE (column_names).

Give it a read if you want to refine your SQL skills!

What’s going on at Paradime?

(Blog) Stuff we shipped #8 by Kaustav Mitra (CEO, Paradime)

Every month, Kaustav shares new features that we've rolled out for our customers. In this blog, you'll discover various new Paradime features, including:

• Search experience in Code IDE - Search and replace anything and everything within your dbt project. Check out the hands-on video tutorial.
• Paradime Live v2.0 - See the data behind your code in real time as you create and update your dbt project.
• Outgoing Webhooks - Seamlessly create outbound webhooks to dozens of applications for your scheduled dbt runs.

(Podcast) Somewhat Analytics Podcast - Ep #5 w/ John Chen

When Kaustav Mitra and I launched the Somewhat Analytics Podcast, we set one fundamental rule: Only interview guests who genuinely intrigue us.

Good news: John Chen gets a big thumbs up from both of us! ????

He's a Sr. ML engineer at Meta and also an NBA analyst. In this conversation, we delve into various topics, including:

?? Being analytics-driven in the NBA demands filtering out the noise from the data (Don't get overly excited about a single outstanding game from a player!)

?? Alex Caruso and Jrue Holiday are the dream players for analytics enthusiasts. Historically, they boast impressively high +/- and significantly contribute to wins.

Hands on Paradime Tutorials:

At Paradime, we take immense pride in the quality of our documentation, tut why stop there? Every week, we produce hands-on, step-by-step video tutorials to demonstrate precisely how to implement our product. Below are some of the tutorials we've unveiled recently:

???**How to trigger Paradime Schedules through Airflow** - Delve into connecting your dbt schedules in production using Airflow. For added guidance, we've also provided a GitHub repo with comprehensive instructions to assist you in implementation.

????*How to make CI/CD work suck less* - This tutorial elucidates how to run CI/CD within Paradime, highlighting two essential CI/CD features:

• Lineage Diff - When initiating a pull request, Lineage Diff detects alterations to columns in your dbt models, subsequently generating a report on all affected downstream entities (e.g., Tableau dashboards). With Lineage Diff, you'll gain confidence with every pull request you merge into production.
• Turbo CI - Turbo CI facilitates the testing of code modifications to your dbt project before merging them into your main branch, ensuring that your production schedules remain unaffected when pull requests are integrated.

(Customer Story)Tide’s #1 favourite thing about Paradime

Tide has been a valued customer of paradime.io for quite some time. Their feedback has driven us to develop and refine numerous features on our dbt? native operating system, including:

?? Multi-workspace - Own and manage multiple dbt? projects, each with its own repo, users, data warehouse, production schedules, and more.

? Bolt Schedules - Build and maintain production schedules within a .yml file located in your git repository.

?? Webhooks - Transmit data about production schedules to virtually any SaaS tool.

That's all for this edition of the Somewhat Analytics Newsletter… See you next month!