Fresh Reads

I have been without my Mac for 5 days: no coding, no AI, no nothing. Out in the wild. After these 5 days, here is everything that I missed out while off screen.

This is the last guide you need to write great code with AI. And it's nothing impressive, but mostly a series of high-level steps that, if done correctly, guarantees success with AI coding.

IndexedDb in effect is finally here! A full type-safe and local-first enabled IndexedDb module is out in effect v4 beta, and it finally unlocks using IndexedDb for real. Get started here.

Another experiment with AI this week, and this time it went extremely well. I ported Bevy (game engine) to TypeScript, with the patterns and type safety of effect. All AI. This is why it went so well.

In an attempt of going from requirements to full implementation fast with AI, this week I tried an experiment: all-in AI specs and code. Well, it failed, bad, and rather fast.

I have been running full AI-mode on a few projects for a few weeks now. I just now paused to take a look at the overall code, and the situation is bad. Here are some ways to fix and prevent this situation.

You probably cannot write code as fast as AI. But you don't have to. Even if you are not typing every word, it's still your job to ensure a working final product. And there are a few strategies you can use.

From State of React it looks like only 10% of people used XState. I was one of those many moons ago, and after I discovered it, I never looked back. And the reasons are practical, let me tell you.

Effect v4 beta is out. It's time to start migrating everything to it right away, of course. I already started. Here is how it went, what you should know, and what are the improvements.

AI gets better, faster, smarter, and more precise. But some hard limits (still) exist. And that's the value you bring as a Software Engineer, as you work with AI.

My AI workflow changed (again): from a single AI running the show, to multiple AIs in parallel. I became officially a conductor of AIs, with effect acting as a guard to check the proceedings.

What's the lever that maximizes each step in your journey to mastery? It's called Intent. It removes uncertainty, increases focus, and skyrockets confidence and motivation. Let's see how.

How does the path look like from good to better? Turns how the shape is always the same, and it repeats in cycles. But it's not linear, it's not constant, and it involves a lot of down periods.

With AI, streaming is not optional anymore. But streams are complex, aren't they? Well, not really with Effect. This is how I use Effect to make AI requests with streams, parse the result, and wire all together server to client.

TypeScript API? Well, use Effect. And since Effect is just TypeScript, it just works everywhere. I have been implementing an API with Effect on Cloudflare, and it all works, great. Here is my setup.

Many are introduced to effect for error handling and dependency injection. Those are huge wins, but effect goes way beyond that. Let me show you a couple of concrete examples.

If 2025 was not enough, 2026 prospects are even more groundbreaking. A few major release will shake my stack once more: effect v4, xstate v6, TanStack Start, Drizzle v1.

Effect is everywhere in my codebases, full stack, back and front and scripts and whatever. A few core module are always there, and carry most of the logic. Here they are, and how I use them.

When should you move fast and work hard, and when should you take a step back and plan? Neither precision nor speed in isolation is the solution, each calls for the other, in a structured process.

State management looks easy enough, until it doesn't anymore. That's why you should use XState. And if things are really complex, you will need a full actor system. Let's see how.

The same guardrails we created for humans when writing code must be applied to AI. First check is types (Strict Type Safety), then linting, and testing as the ultimate step. A few notes.

When code is written mostly by AI, how do you stay relevant as a software developer? Well, you switch from pure development, to architect and engineer. This is how.

I recently deployed a full stack effect app, mostly written by AI, super fast and super type-safe. Effect is a super power, and the AI can leverage it. Here is how.

Coding with AI is becoming more and more a necessity. The new challenge is making sure the AI writes type-safe and maintainable code. Well, it's possible, here is how.