ai-architecture
6 articles with this tag.
The Two Failure Modes Nobody Formalizes
Every AI team can tell you what happens when their system works. Ask what happens when it fails silently, and the room goes quiet. Two disciplines close that gap — and almost no one has named them.
Five States, Not Two — Why Most AI Tools Treat Their Users Like an Afterthought
Most builders design for success and bolt on an error page. I designed five explicit states. Here's why the failures are the product.
We Published Our AI Rubric Calibration Data. Everyone Told Us Not To.
The AI judge scored 60% of responses at Level 4. Manual review put the real number at 15%. We published the gap.
Our AI Judge Went Down. Nobody Noticed.
The system didn't crash. It succeeded at writing to the wrong place — for weeks. Here's what I built after.
Every Input Is Hostile — Even When It Comes From Your Own Team
One malformed JSON record took down an entire notification pipeline. The input wasn't an attack. It was a database migration.
I Built 20 Autonomous AI Agents. Then I Learned Why Evaluation Has to Come First.
It cost me 14 failed tasks and a system I had to rebuild from scratch. Here's what changed.