Less pretending. Fewer masks.
Whether full-time or solo, let the work finally serve you.
Teams matter. Products matter.
But none of it makes sense
if it doesn’t move you closer to your own goals.
I wish this for myself — and for you.
#NewYear2026
> be AI company
> scrape the entire internet
> ignore licenses and copyright
> download millions of pirated books
> use them to train frontier models
> spend tens of millions buying physical books
> cut off their spines
> scan every page
> destroy the originals
> lecture everyone else about intellectual property
> we are the good guys
"For your own safety, we will have to revoke all your rights. We wouldn't want you to accidentally do something harmful to yourself."
The system appreciates your understanding.
Polish companies are accelerating digital transformation.
People are absorbing the cost.
85% report lower effectiveness due to information overload.
40% fear AI may replace their roles.
Change moves fast.
Humans don’t scale at the same speed.
#futureOfWork
I’ve spent years reading systems from the inside out. Good tools don’t guess — they reduce noise.
AI fits best when it helps you see structure, not shortcuts.
#softwareEngineering#LongTermThinking
One of the genuinely wonderful things AI does: writing regex.
What used to be a dark art is now a quiet collaboration.
Less frustration. More focus.
I’ll take that.
For years, programming was my escape.
A safe place where flow took over.
What happens when it’s no longer safe?
Something is shifting.
#SoftwareEngineering
I’ve seen tools come and go.
The good ones don’t replace judgment — they sharpen it.
AI feels like that kind of tool. Quiet. Useful. Earned.
#longTermThinking
Turns out Polish might be one of the best languages for AI chats.
Beautiful.
Now the question is what we’ll actually do with it.
Probably build things… for others.
Feels like a familiar story.
I wish embeddings and vector search were around 20 years ago.
It would’ve saved countless hours fighting categories, names, vendors, and wholesale chaos.
Different tools. Same problems.
Beautiful times. #devThoughts
Is there already such a thing as an "external hardware LLM" like we have external hard drives?
Instead of having to run/maintain a local model, I want an inference machine that I can just plug in and point my prompts at. Single GPU, maybe a few in parallel.
Who's building this?