Hello, world: starting to write

I've been meaning to write more for a while. Mostly short notes on things I learn while building: the kind of stuff I usually scribble in a private notebook and never look at again. This space is the public version of that notebook.

What you'll find here

Roughly three kinds of posts:

• Tutorials: step-by-step walkthroughs of things I've built or learned.
• Notes: short, focused write-ups on a single idea or problem.
• Postmortems: what went wrong, what I'd do differently, and what stuck.

Why bother?

Writing is the cheapest way I know to find holes in my understanding. If I can't explain a concept cleanly to a stranger, I probably haven't earned it yet. So this is partly for me, and partly for anyone walking the same path.

A small example

Most posts here will involve some code. The reading layout supports inline things like tokenizer.encode(text) as well as code blocks:

from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("eolang/SW-v1")
model = AutoModel.from_pretrained("eolang/SW-v1")

inputs = tokenizer("Habari ya asubuhi", return_tensors="pt")
outputs = model(**inputs)

That's it for now. More soon. Thanks for reading.