Tech & AI
How artificial intelligence is reshaping industries, products, and the way we work, from a builder who uses it as infrastructure, not a feature.
11 articles
Swipe Right on This: Dating Apps and LLMs Are Running the Same Playbook
Dating apps and AI chatbots both cut you off before it gets good, but one is managing infrastructure and the other is managing your dopamine.
ChatGPT Has a Human Problem. And That's Not a Compliment.
I asked ChatGPT about Real Madrid and it confidently told me about a manager who'd already left the club, and that one moment explains everything wrong with how we're trusting AI right now.
Is Microsoft the New Nokia?
Everyone thinks Google is losing the AI race. I think the more interesting question is whether Microsoft is quietly becoming the next Nokia.
I Hired Two AI Developers. One Is a Rocket. The Other One Checks the Wiring.
Managing Claude and Codex as a solo founder felt uncomfortably familiar, turns out scaling AI agents has the same team dynamics as scaling a real operations team.
You Don't Need to Code Like a Developer, You Need to Think Like a Product Owner
I didn't become a developer. I became the one-man product owner, scrum master, and stakeholder who finally had an AI team that could actually build what I was describing.
AI and the Logistics Layer Nobody Talks About
Every AI demo shows the glamorous output. Nobody shows the warehouse chaos underneath it.
The Intern Who Outworked the Veteran: Why Codex Fixed in 10 Minutes What Claude Couldn't in 2 Hours
I gave up on a bug, handed it to OpenAI's Codex almost as a joke, and watched the new kid solve it in 10 minutes with evidence, logic, and a slightly annoying smirk.
Claude Opus 4.7 Is Here. Mythos Is Not. Here's the Difference.
Anthropic just dropped Opus 4.7, the most capable Claude yet, while quietly confirming that Mythos, its successor, was restricted over global cybersecurity concerns.
I Hit My Claude Limit on a Tuesday. Then I Met Gemini.
I ran out of Claude mid-week and decided to give Gemini Pro a real shot, what followed was like switching from a five-star hotel to a party hostel where the staff keeps pointing you to the map.
I Asked an AI to Fix My Build. It Took Two Hours and Taught Me Something Uncomfortable.
A post-mortem on a two-hour AI debugging session that should have taken ten minutes, and what it says about how I work with these tools.
I Caught ChatGPT Mixing Cyrillic Into German Mid-Word. Here's What Actually Happened.
ChatGPT wrote a German word with Cyrillic characters embedded inside it, and then spent six attempts failing to correct it.
Learn the mechanics
Learn
Send a read-only agent first
One agent spent three hours chasing a build error. A second agent read the migrations against the query code in two minutes and found the real bug. The lesson isn't about which AI is smarter, it's about audit-first workflows.
Learn
How to split work across two AI agents without merge conflicts
When two AI agents work on the same codebase in parallel, file-level collisions are inevitable without a deliberate coordination pattern. Protected lanes and explicit ownership boundaries solve this without requiring real-time communication.
Other topics
Finance & Investing
Markets, crypto, private equity, and the psychology of capital. Personal opinion, not financial advice.
Cars & Mobility
Used markets, EVs, depreciation, city movement, and what vehicles reveal about status, utility, and how people actually get around.
Health & Longevity
What burnout taught me, and everything I've learned since about the gym, nutrition, sleep, and the science of staying sharp for decades.
Philosophy & Ideas
Big questions, slow thinking, and books worth losing sleep over. The kind of ideas that change how you see everything else.