Is a passing test suite enough to establish that code is correct?
The question A common position, especially now that most code is written by AI agents, runs as follows: write tests with enough coverage, and then it does not matter what […]
What’s the shortest program for which we can’t tell if it stops?
Here’s a 10-line Python program. Nobody on Earth currently knows whether it ever stops. The halting problem is the question of whether, given a program, we can tell if it […]
The Morse code Grok heist for dummies: who robbed whom and how
The Heist On May 4, 2026, someone stole about 150,000 dollars by sending a tweet. How is it even possible? Where did the money come from? The answer involves Grok, […]
New excuse in town: “my subagent did it”
AI agents are known for making mistakes in multiple ways. Sometimes they just make things up, aka hallucinate. Sometimes they speculate. Sometimes they are unaware of major developments that happened […]
How to lose your folder: the dangers of storing data in the file system
The Incident A few days ago, we upgraded our server from one type of storage to another. Suddenly, most of our data disappeared from view. It still existed on disk, […]
Can Comrade Claude code in an Ancient Soviet Language?
Can an LLM code in a language it was not trained for? Spoiler: kind of. It is not very fast and not very efficient, but if you give the LLM […]
Daylight savings time is evil
Daylight Savings Time is evil and should be outlawed (Putin’s Russia has already done that! 🙂 ). I am building directing Cursor to build a dashboard and it’s really a […]
Vibe coding: code review still required
Another AI/vibe coding battle story which makes me hard to believe it is possible to create reliable products through vibe coding alone without code review. The AI emulates humans too […]
IKriv Certs 1.0 – SSL Cert Expiration Checker
When LinkedIn announced they will no longer support email alerts when the certificate expires, I quickly vibe-coded the solution, but it took me a while to “productionize” it so I […]
Example of LLM confusion over a word with multiple meanings
In our system we can “hide” a user conversation in two ways. There are internal “service” conversations that are technically owned by a user, but are not supposed to be […]