Posts

There’s an agent design pattern with an uncomfortable premise: the agent is the orchestrator, and you are one of its tools. It goes by a few names — human-as-a-tool, human-as-tool, sometimes lumped under “human-in-the-loop.” But it’s worth separating the two, because the inversion is the whole point. Human-in-the-loop vs. human-as-a-tool Human-in-the-loop keeps the human in charge. The agent proposes, the human approves. You’re the supervisor signing off on the intern’s work before it ships....

When you run a team, you don’t decide who to keep based on their resume from two years ago. You watch what they actually ship on your work, and you promote the ones who deliver and let go of the ones who don’t. I want the same thing for the models I use as agents. And public leaderboards don’t give it to me. Why public benchmarks don’t tell you what you need A leaderboard tells you how a model did on someone else’s tasks, scored on someone else’s rubric, often on problems that leaked into training data....

This is an opinion piece. My read on the news, not reporting. Disagree freely. Last Friday (June 12, 2026), the Trump administration forced Anthropic to pull Mythos 5 and Fable 5 — three days after launch. The official reason: a “national security” risk over a minor jailbreak. The real reason looks simpler: if we can’t control who uses it, nobody gets it. I think that’s a mistake. Here’s why....

You probably think of GitHub Actions as a CI tool: push code, run tests, deploy. But strip away the CI framing and what’s left is something more interesting — a free virtual machine that boots on a cron schedule, runs whatever you want, and shuts down. No VPS bill. No Raspberry Pi humming in a closet. No “did my home server lose power again?” Every scheduled run gets a fresh Ubuntu VM with 4 vCPUs, 16 GB of RAM, Python, Node, Docker, and internet access....

Running models locally with Ollama is great for privacy: nothing leaves your machine. But it comes with a hard ceiling — your model only knows what it was trained on. Ask it about something that happened last week and it confidently makes things up. The usual fix is to plug in a web search API (Tavily, Serper, Brave). The problem? You’re back to shipping every query to a third party with an API key tied to your name....

Why Run Kali in a Container? Let’s be honest: modern workflows — especially if you’re experimenting with AI agents like Agent-Zero or automated LLM pipelines — demand environments that are programmatic and ephemeral. An AI agent isn’t going to navigate a GUI-heavy VM; it’s going to spin up a container, execute its task, and vanish. The container approach offers: Isolation — your host OS stays clean. No rogue binaries, no broken dependencies....

I had a different post queued up for this week. It was going to be a practical post about Immich: how I finally moved my photo library away from Google Photos, how the timeline view works, how face recognition feels, how “on this day” memories hit differently when they are not attached to an advertising profile. That was the plan. Then Google signed a classified AI deal with the Pentagon....

Ally Piechowski recently shared five Git commands she runs before reading any unfamiliar codebase. Some were already in my toolkit, some were new to me, and all of them were useful. I liked the idea enough that I added my own five, so this became a practical top 10: five from Ally, five I keep reaching for myself. None of these commands tells the whole story. But together they give you a fast read on churn, ownership, risk, history, and maintenance patterns before you open a single file....

When I’m in AI Studio mode (rapid prototyping, lots of experiments), I want to spend my time on prompts, evals and breaking the code — not on re-learning yet another SDK. That’s why I use aisuite for most of my AI projects: One client API across providers Switching models is usually just changing a string like openai:gpt-5-mini → minimax:MiniMax-M2.1-lightning or even local ollama It stays close to the OpenAI-style shape, so it’s easy to adopt Lately I use quite a bit Minimax for coding tasks because it hits a great ratio of price vs quality....

The Trading Parallel Chess and trading share a fun truth: you can spend hundreds of hours practicing and still lose. In trading, it’s capital. In chess, it’s ELO. Both require deliberate review of past mistakes to improve. Traders keep journals. Chess players should too. I’ve found that Obsidian—a tool I already use for notes—works perfectly as a chess game database. With the Web Clipper extension and a couple of templates, every game and blunders becomes a searchable, annotated note with an embedded board....