Blogs

Notes on AI tools, agent workflows, and what actually ships in the real world.

Two feeds, same focus: practical AI work, reviews, and field notes.

Substack Medium

Skills Are Basically MCP Servers in a Different Jacket

Now, to be fair, there are differences. Real ones. They just don’t make them fundamentally different entities. They make them different delivery and orchestration choices . Difference 1: Context economy is handled …

substack · abvx.substack.com

The End of “Prompting” and the Rise of Agent Engineering

Claude Code, Codex, MCP, multi-agent swarms, context layers, and why the real competitive edge is no longer the model — it’s the system around it There was a brief, glorious period in AI coding when people thought the ga…

medium · abvcreative.medium.com

Lean Vibe Coding

I think I’ve just stumbled into a useful term, so I’m claiming it here first: Lean Vibe Coding And the funny part is this: Toyota’s production system, lean manufacturing, Goldratt’s constraints th…

substack · abvx.substack.com

The Great AI Slop Paradox

Why audiences can’t spot “AI content”… right up until they can — and what platforms, creators, and Hollywood are doing about it. A funny thing is happening to the internet. Not the usual funny thing, like a billionaire b…

medium · abvcreative.medium.com

AI FOMO Is the New Normal (And It’s Not Going Away)

For the last few months I’ve been living in a permanent state of AI FOMO. Not the cute kind. Not the “oh no, I missed a party” kind. The industrial kind. The kind where you wake up thinking: “I sh…

substack · abvx.substack.com

The 2026 Internet Utility Belt: 20+ Tools That Actually Help (And a Few That Will Ruin Your Life)

A practical field kit for creators, builders, and operators — plus the security hygiene to survive all the “free” shortcuts. The modern internet is now a hardware store run by caffeinated raccoons. There are tools everyw…

medium · abvcreative.medium.com

The Last 100 Jobs the Robots Will Struggle to Take

A very smart person once suggested a mildly evil way to read AI benchmarks. Not: “Where does AI beat humans?” But: “Where do humans still win — and why?” Because benchmarks are basically the…

substack · abvx.substack.com

The Year We Stopped “Using AI” and Started Living Inside It

Why the next wave isn’t about smarter models, but about who gets the leverage — and who gets left behind. There’s a particular kind of moment in human history where the headlines are technically visible… and yet nobody b…

medium · abvcreative.medium.com

Five Slightly Unhinged Claude Code Tricks (That Quietly Make You Twice as Dangerous)

Let’s be honest. Most advice about coding agents is either: “Just prompt better.” (Brilliant. Like telling someone to “just be taller.”) “Build a framework.” (Yes, sure, let me s…

substack · abvx.substack.com

Memory Is the New Bottleneck: Why “Second-Wave” AI Agents Live or Die by What They Remember

Why the next generation of AI agents won’t be won by bigger models — but by systems that remember, update, and forget on purpose. A long-read review of Rethinking Memory Mechanisms of Foundation Agents in the Second Half…

medium · abvcreative.medium.com