Lately, as I build AI agents, I’ve found that no matter what tools, frameworks, or scaffolding you use, a solid architecture keeps in mind a language model (the brain), tools (for its actions), memory (what it knows), channels (for communication), and loops and triggers (to keep it acting and reacting).

It’s dangerous when you found yourself can’t verify the outputs from AI.

Some say vibe coding produces nothing but useless stuff — that nobody uses any of it, that it's all people impressing themselves. But writing code was never the bottleneck. It's maybe 20% of the job. Once you know what you're building and how it should fit together, the code mostly falls out. Vibe coding took the easy part and made it easier. Good products that actually get used have always been rare. The barrier dropped, so far more gets built, and of course you see more junk. Plenty of people who shouldn't be shipping are now shipping, so the hit rate probably went down. But in absolute terms there's more good work than before, and the good work ships faster. And whether people use your product comes down to whether it gives them something they want. Not the framework. Not the architecture. Not whether a human typed it.

AI isn’t perfect. But neither are we — and perfection was never the point. The real goal has always been getting things done well enough. AI just helps us learn faster and clear that bar across more domains. Stack up enough “good enough,” and that’s how productivity actually improves.

"Coding is solved" — Anthropic's Claude Code playbook makes it almost plausible.What they don't say loudly: the engineering didn't disappear, it moved. You maintain CLAUDE.md, build MCP servers for your org, package expertise into skills.Different work. Still work.

I think the quality of AI-generated code is equal for frontend and backend. The reason people think AI is disrupting frontend more is that visual elements are intuitive — you can actually see the result. You have an idea, and you can vibe-code a frontend demo right away. But nobody would vibe-code a bunch of APIs for no reason.

I was asked what I most expected AI to accomplish in 2026. Though I pointed out something very likely to be developed this year, I couldn't think of anything I truly EXPECT. Yeah, I read a lot, I learn a lot, I have taste, and I can pick up any new tech — but I should be more opinionated. So that at some point I'll think, "Oh, I really want this. I expect it. And if no one does it, I'll build it myself."

Always have ai agents working in the background…

Everything goes CLI and a set of skills now.

It is never too much trouble to treat your AI agents well. - Give an appropriate amount of and tailored context to them - Build tools for agents and tell the agents how to use them - Be consistent and comprehensive with the docs and domain knowledge - Give them access to your work and let them test and validate themselves - Give them specific and doable work and always use "please"