Behind the site

Why lostcontext.ai?

The name has two meanings — and both are on purpose. On the mission, building from the inside, and what should grow here.

This is the first piece here — a statement of intent, before the actual articles arrive.

Two meanings

"lostcontext" means two things, and both are on purpose.

Technically, the name describes a real phenomenon. Every large language model works with a context window: the slice of a conversation it can process at once. Fill that window and older parts drop out — the model loses the thread. The gradual decay in quality that follows even has a name in the field: context rot.

Socially, the name means something else. The public debate about AI keeps losing its context, too. It tips between salvation and doom and skips the concrete middle: what these tools can really do, where they help, where they fail, and why. That's the context I want to hand back.

From the inside, not the sidelines

I don't write here as a spectator. I build applications — and I use the very AI tools I write about while doing it: daily, in real projects, with real mistakes. That vantage point is the heart of lostcontext.ai. Not "an expert says", but "here's how it went when I built it".

What grows here

Two things. First, articles that show how these systems work and where their limits are — calm, concrete, no hype. Second, a Lab of small interactive pieces that make an abstract idea something you can touch.

What won't happen here: no scaremongering, no superlatives, no numbers without a source. Where I can't back something up, I say so — or leave it out.

Transparency

Right now this site earns nothing and costs no one anything. Should that change — through restrained advertising, say, or the odd recommendation link — I'll announce it here, openly, before it happens. No hidden tracking, no bought opinions.

An invitation

If you've read this far: glad you're here. This place grows slowly and on purpose. Come back — or write to me if something's on your mind.

/compact — the essentials, if context is running low:

"lostcontext" means both: language models lose context (context rot) — and so does the public AI debate. This site is written by a practitioner from the inside, calmly and with no numbers without a source. What's growing here: articles, and a lab that makes abstract ideas tangible.

An unhandled error has occurred. Reload 🗙

Rejoining the server…

Rejoin failed… trying again in seconds.

Failed to rejoin.
Please retry or reload the page.

The session has been paused by the server.

Failed to resume the session.
Please retry or reload the page.