Skip to content
Sandro Maglione

NewsletterIssue #1523 min read

What should you do when code becomes free

The skill of writing code will become obsolete eventually, we are getting there faster and faster. And if writing code has not much value anymore, you need to shift your focus. Here is what I am doing.

Contents
  1. You are good at code, but code is no more
  2. Creativity and tooling
  3. Scope and personal projects
  4. Game development with ECS

Programming is not the main skill anymore 🤷🏼‍♂️

And it's time to adapt, for real 🫵

If writing code is not a thing anymore, as a programmer you need to branch to a new set of skills.

Here is how I am navigating the shift 👇

You are good at code, but code is no more

We reached the inflection point:

No one expects you to write code anymore 🙌

In fact, if you write code is suspicious, a sign you are not adapting to what coding is becoming (exceptions apply, as always, of course 💁🏼‍♂️).

In my experience, this is slowly getting to the point where coding is not a "skill" anymore:

If your only skill was writing the code, you need to pivot, soon 🙌

I am branching into other skills now:

  • Management (of agents)
  • QA
  • Observability
  • Performance
  • Product

I actively need to practice, engage, and learn more about those.

Creativity and tooling

If code is not a bottleneck anymore, and an agent can implement anything, you need to expand your vision.

Anything that you think may be somehow worth having, no matter the scope, it's most often worth a try (i.e. prompt) 🤔

Need better devtools? Automate a manual task? Any kind of new feature?

It's all cheap, from a code perspective. Any time you find yourself doing something manually, it's worth questioning why an agent could not do it.

Most time, the agent could, given the right access 💁🏼‍♂️

You still need judgement to decide if a feature is worth adding or not. But, even then, the cost of removing it (cleanly) is going down, enough that it's often worth trying adding it anyway 👀

Scope and personal projects

Another mental shift I had:

The scope of personal projects increased by orders of magnitudes 🚀

What previously was a "multiple months project", it's now a weekend prompting away 🙌

Anything that you consider interesting to have, for you, no matter the scope, it's all withing reach.

Game development with ECS

Realizing the scope availability, I picked up (again) a new/old project, and released it: @typeonce/bevy-ts

It's a full ECS (Entity Component System) "game engine" in TypeScript. It's what I wanted for game development: type safety, agents first, composable, fast.

And it's all agent-driven: the target is ease of development for agents, debugging, type safety 🪄

This would have been a huge project (years) before. Now, a few reference, preferences, prompts, and it works.

Since code doesn't matter much, the focus is all on gameplay and game-feel.

So much so, that to get better I am actually playing more games, instead of studying about game development principles 💁🏼‍♂️

There is still time to shift your focus.

Look around for anything that previously was outside your role, and now it's slowly getting within the responsibility of software.

No matter the scope.

See you next 👋