the journal

AI in production.

Field notes from running agent fleets in production — what survives past the demo, and how dev teams become agentic operators.

Latest

Is Jev a scam? What TypeSafe AI's System One model is, and what happened when we ran it

Jev is a real, hosted, closed model that returns typed decisions with a probability vector. Its 193x / 445x marketing numbers are unproven by anyone, including TypeSafe. We ran jev-1.13.0 on Swiss bookkeeping data: accurate and calibrated on 54 real accounts, confidently wrong on a 10-item cohort.

23 Sept 2026Agency

The $2 trillion question
Agency14 Aug 2026

AI in production.

Is Jev a scam? What TypeSafe AI's System One model is, and what happened when we ran it

The $2 trillion question

Code is still law

AI agents in production: the four operating problems that decide it

Your AI agent didn't fail the deploy. It stopped itself.

Your failing agents waste most of their tokens after the warning signs

A keyword search beats most AI agent memory

How to observe AI agents in production

Your most accurate agent setup is the wrong one to ship

When your agents fail, can you tell which one did it?

What AI agents actually cost, and where the money goes

Your agent's success rate is one lucky run

How to make AI agents reliable in production

Your AI agent says it succeeded. Don't believe it.

Context engineering for AI agents: less, but the right context

More context is making your AI agent worse

Why your AI agent can't correct its own mistakes

Our agents are our first users. So we interviewed them, and only believed the logs

The EU AI Act reaches your non-EU software team, and the delay doesn't save you

AI and Swiss secret professionnel: the three DPA terms that decide whether the secret survives

Get the notes