Navigate
Technical Writing
Deep dives on the systems, patterns, and engineering decisions that let AI teams ship with confidence.
3 Jul 2026
6 min read
I have been building agents without a runtime. It works until something breaks and you have no idea why. Here is the runtime model I wish I had from the start.
23 Jun 2026
6 min read
I built an autonomous remediation agent. It worked for two weeks, then it restarted a production database pod during a backup window. Here is what I learned about where agents actually belong.
6 Jun 2026
6 min read
I ran the numbers on my own setup. Self-hosted AI is not always cheaper, and the hidden costs are not where you think they are.
28 May 2026
7 min read
I almost sent customer data to a public LLM API because my pipeline was not checking the data flow. Here is the security model I use now.
22 Apr 2026
5 min read
I lost a day's worth of data because n8n does not replay state. That is why I now run Temporal for anything that matters.
7 Apr 2026
6 min read
I gave an AI agent kubectl access once. It deleted the wrong namespace. Here is why I now believe MCP servers are the only safe way to let agents touch infrastructure.