Thoughts on AI engineering, responsible automation, and building systems that matter.
At my first startup, documentation meant a README with three bullet points and a prayer. Then I joined a German enterprise and discovered that documentation is not overhead - it is infrastructure. That shift changed how I build everything.
Read more →
The bug crashed production every Tuesday at 2 AM. Only Tuesday. Only 2 AM. The logs showed nothing. The monitoring showed nothing. Three days later I found it - and the fix was not code.
Read more →
My team needed a live dashboard showing CI/CD pipeline status across 40 repositories. I reached for WebSockets. Then I stopped, thought about it for five minutes, and chose something simpler.
Read more →
Every developer on my team uses the same AI setup now. Not because I mandated it, but because I built skills so useful they spread by word of mouth. Here is what I learned from building a shared configuration that actually works.
Read more →
I had a production hotfix, two feature branches, and a code review all due by end of day. Stashing was not going to cut it.
Read more →The codebase had 847 suppressed warnings. I know because I counted them on a Friday afternoon when deployment failed for the third time that week.
Read more →
VSP-727 appeared in my Jira queue at 9:14 AM. By 9:29, the fix was in production. Here is the pipeline that made that possible - and why I built every piece of it myself.
Read more →
I wanted an AI assistant that could check my email, manage my calendar, generate leads, send outreach, and monitor the web - all while I sleep. Six months later, she runs operations across 7 countries.
Read more →
The third time a broken test made it to the remote repository because a developer forgot to run the test suite locally, I decided to solve the problem permanently.
Read more →
The first time a lawyer asked me "but who is liable when the AI makes a wrong decision?" I realized I had been building systems without understanding the world they operate in.
Read more →At 6 AM on a Tuesday, while I was still sleeping, my AI agent found a lead in Melbourne, generated a personalized website mockup, crafted an email referencing their specific industry, and sent it. By the time I woke up, the prospect had replied.
Read more →Last week, I watched a junior developer generate an entire authentication system in four minutes using AI. He could not explain how password hashing worked. This is the vibe coding problem, and it is bigger than people think.
Read more →
My first week at the new company, I tried to paste a code snippet into ChatGPT. My colleague physically reached over and closed my laptop. "We do not do that here." That was my introduction to real AI governance.
Read more →
The first time I had to fire someone, I rehearsed the conversation for three days. They were a friend. They were also three weeks behind on every deadline. Nobody prepares you for that at fifteen.
Read more →
The CEO called me into a meeting room with no windows. "We are losing customers faster than we can hire support agents." That conversation changed the next year of my life.
Read more →
I needed to build a CLI tool that could orchestrate Docker containers, manage Git worktrees, allocate ports, and run a 9-stage pipeline in under a minute. I evaluated five options. The winner surprised me.
Read more →
I used to start every morning with three browser tabs, three inboxes, and a growing sense of dread. Now an AI agent reads them all before I wake up. The journey from inbox chaos to automated calm took six months and changed how I think about attention.
Read more →
The first version of my voice assistant sounded smart and felt terrible. You asked a question, waited four or five seconds, then got an answer. Technically correct. Emotionally dead. Getting that latency below two seconds changed everything about how I design AI systems.
Read more →
There is a healthy way to code with AI, and there is a dangerous one. I have worked in both modes. One makes you faster without stealing your judgment. The other gives you velocity up front and a mess later. Learning the difference changed how I build software.
Read more →
AI can write code quickly. That does not make the code safe, maintainable, or production-ready. Over time I built a testing workflow that treats AI output like a fast junior teammate: useful, productive, and never above verification.
Read more →