What actually survived

Last time I talked about automations that died.

But what actually survived?

Here are three things that really work. Still do.

A bot for language practice

The simplest one of all.

I send it an English text. It takes it and puts it into a table. Counts the words. That’s it.

No AI, no magic. Just an archive of what I wrote.

From time to time I open the table, reread the texts, and notice my own mistakes by myself, without a teacher.

I try to write at least 500 words a day.

It works precisely because the bot does not try to be smart. It just makes sure I do not lose what I wrote.

It takes RSS feeds from Google Alerts based on specific keywords.

AI looks at every piece of news and decides whether it is relevant or not. Relevant ones go into a table.

When enough items accumulate, it takes the headlines and scores each one on a 10-point scale based on how useful it is for our channel.

The result is a list with ratings.

I open it, look at everything above 7, and start reading.

No human is needed in this chain. The bot runs on its own.

A robot monitors hundreds of channels by keywords and collects everything I need in one place.

I use it for three things:

It sounds complicated, but in reality it is just a filter instead of scrolling through everything myself.

They do not try to do anything complex for me.

They remove the mechanical part: collect, store, filter.

The thinking still stays with me.

Maybe that is exactly the boundary between things that survive and things that die.