On Building AI Understanding and Automation Muscle (With 18 Random Problems Solved With AI)

Published on January 29, 2025, filed under (RSS feed for all categories).

If you’re like me, you’re also optimizing your use of AI in development, distinguishing your capabilities from AI (cf. 8 AI Tips for Web Developers), and improving your routine to build strong automation muscle (i.e., to instantly recognize and seize opportunities for automation).

For me, deliberate, continued use of AI in my development work has helped me in all three areas, to a degree where I have a better idea of where AI helps me and the organizations I work with, where it can’t and won’t “replace” me, and at which I leave fewer and fewer options on the table (and lose time doing so) when there’s something I can automate using AI.

In fact, before we start: Throwing AI at anything that looks automatable is a great use of AI, to solve the problem, and to develop said muscle.

A Great Number of Random Projects That a Person on the Internet Came Up With to Solve the Problems That May or May Not Be Remotely Like Your Problems

I’m sharing the following 1) to provide some examples of AI-solved problems (in the small), 2) to maybe help immediately solve the problem, if it’s also yours and the solution is public, and 3) to be kindly notified if I have missed something important.

  1. Use AI to write a shell script to find the most frequently used authors and publishers on Frontend Dogma, to have easier access to that information and support Frontend Dogma’s Mastodon-first policy.

  2. Use AI to write a shell script to identify underused and dormant (commented) tags on Frontend Dogma, to make it easier to populate or enable such tags.

  3. Public: Use AI to set up Jest tests and integrate tests in CI for Imagemin Guard, to improve robustness and speed up releases.

  4. Public: Use AI to build a script and npm package to identify obsolete HTML elements and attributes, to enable and encourage that work.

  5. Use AI to write a Python script (regularly invoked by cron job) to post web development glossary terms of the day to Bluesky, not to feel forced to pay for what also seemed to be overpriced solutions.

  6. Use AI to write a Python script (regularly invoked by cron job) to post Frontend Dogma content to Bluesky (likewise complementing publicly available services for other networks).

  7. Use AI to write a Node script (regularly invoked by cron job) to check on automated posting to Mastodon, to be sure aforementioned automation is working reliably.

  8. Use AI to write a Python script to automate the selecting, committing, and exporting of new posts to be published on Frontend Dogma, to improve and speed up content management. (Given the huge number of entries linked on Frontend Dogma, with me still making an effort to scout thousands of articles from the 2000s and 2010s, these optimizations save a ton of time.)

  9. Use AI to improve tagging efficiency in Frontend Dogma’s front matter generator (e.g., by having it produce a script that manages tag associations), to accelerate content ingestion.

  10. Use AI to write a Python script to post Frontend Dogma webmentions, to test
 webmentions.

  11. Public: Use AI to build a website with Nuxt, to test Nuxt and some more stuff.

  12. Public: Use AI to build a React app using Bootstrap and Storybook, to test all of it.

  13. Use AI to set up D3 to visualize HTML and CSS conformance data, to try D3.

  14. Public: Use AI to build a Remix site, to test some more.

  15. Public: Use AI to write a Python script to sort package.json files alphabetically, because I prefer that over existing scripts like sort-package-json đŸ”€

  16. Use AI to write a Python script to also sort JSON files alphabetically (similar but not equal to package.json sorting), because I needed that for some long JSON files (that GitHub Copilot and JetBrains AI Assistant could not sort themselves đŸ˜Ź).

  17. Use AI to write a shell script to update important forks (maybe overkill, still fun and easy to do), because I had this habit of doing this through the GitHub UI đŸ˜…

  18. Use AI to write a Python script to refactor and add quotes around YAML titles (in Frontend Dogma entries), to limit the risk of issues related to title text contents and to ensure consistency.

This is a snapshot. Personally, I’ve integrated AI in my development cycle and feel more confident about having built a habit of using AI to solve problems and be more efficient. If there’s interest in this, I could share more examples, just as I’d look forward to seeing yours.

AI comes with problems we need to tackle, however, particularly around it being so wasteful, and corporations being able to externalize this waste. And yet in this current system where we need to “earn” our existence and subsistence, not making smart use of AI is also a problem. I hope that in the end, please, we develop more empathy to better watch out for the natural resources we have (alongside watching out for one another).

Was this useful or interesting? Share this post (e.g., on Mastodon or on Bluesky), and support my work by learning with my ebooks!

About Me

Jens Oliver Meiert, on November 9, 2024.

I’m Jens (long: Jens Oliver Meiert), and I’m a web developer, manager, and author. I’ve worked as a technical lead and engineering manager for small and large enterprises, I’m an occasional contributor to web standards (like HTML, CSS, WCAG), and I write and review books for O’Reilly and Frontend Dogma.

I love trying things, not only in web development and engineering management, but also in other areas like philosophy. Here on meiert.com I share some of my experiences and views. (I value you being critical, interpreting charitably, and giving feedback.)