Stay away from the hype and introduce complexity in your systems only if it's warranted by the problem space you have to tackle. Most organizations don't need microservices, use them responsibly.

A good example of using the best tool for the job. Having your whole data analysis pipeline in pandas might not be what you want for performance reason. Very often there's a relational database you can leverage first.

Not exactly new, but easy to loose sight of it when in the hamster wheel. Clearly a good reminder.

Interesting, I didn't know SQLite could have R-Tree indices. This can come in handy for three dimensions or geospatial problems.

Do you ever need to order items stored in SQL? There are ways to do it right and ways to do it wrong. Do it right and pick something meaningful for your case. This article does a good job at listing the typical approaches.