Arnold Moya

[Arnold Moya @_arnold_moya_]

It's worth a try

[Theo - t3.gg @theo]

I've been getting a TON done with Fable today and I'm not hitting rate limits. Wanted to share some tips on how I'm doing that

1. I only use Fable on "high" effort for now. xhigh is token hungry. max/extra is a furnace with worse outputs than lower options imo

2. I taught Claude Code how to use Codex as a fallback for lots of implementation tasks. GPT-5.5 is incredibly steerable, and Fable can learn how to steer it

3. I wrote up a big section in my CLAUDE[.]md on how to prioritize different models for different work when orchestrating workflows and subagents

4. Things that are unnecessarily token hungry (computer use, codebase analysis, etc), I do with other models and report results back to Fable

[Arnold Moya @_arnold_moya_]

@adam_north11 @ClaudeDevs 403

[Arnold Moya @_arnold_moya_]

@mitchellh @cryptotriv Loving Terraform, Vault and Hashicorp. They have made a lot of impact in devs. And yeah, on social networks sometimes being mean works well. Haha

[Arnold Moya @_arnold_moya_]

@thetikiverse @emilkowalski I support your point of view

[Arnold Moya @_arnold_moya_]

@bcherny 2, 4, 5 my focus today

[Arnold Moya @_arnold_moya_]

@james84_ hello

[Arnold Moya @_arnold_moya_]

So the fix for Claude Code’s flickering was basically what OpenCode understood from the start: use a real full-screen TUI instead of constantly repainting a scrolling CLI. Funny how many duplicate “flicker” issues it took before the solution became official.

[Arnold Moya @_arnold_moya_]

@heyalexmoore Need

[Arnold Moya @_arnold_moya_]

Interesting how the frontend performance is improving. Of course, it is the easiest way to show vibe code so model providers are focusing a little more on it.

[Fix @FixlationAI]

GPT-5.6 Sol Tests 🔥

- Frontend Design
- Simulations
- Games and Models

OpenAI have made significant improvements across the board, we are truly in a new era of models...

Do you think it beats Fable? 🤔

[Fix @FixlationAI]

GPT-5.6 Sol Tests 🔥

- Frontend Design
- Simulations
- Games and Models

OpenAI have made significant improvements across the board, we are truly in a new era of models...

Do you think it beats Fable? 🤔

[Arnold Moya @_arnold_moya_]

Sometimes, a single human decision can also save 100k tokens by avoiding an infinite loop of unnecessary agent execution

[Arnold Moya @_arnold_moya_]

Some people are uninstalling their IDEs because they trust AI agents. But the problem is that we lose observability into what those agents produce. Not only code smells, but also quality issues and static checks that are easy to catch in an IDE or CLI tools with visual output.

[Arnold Moya @_arnold_moya_]

Just as architects build scale models, we should be able to inspect the code that is being generated. Well-organized code is easier to understand, improve, and grow

[Arnold Moya @_arnold_moya_]

My smallest relationship was with Fable 5

[Arnold Moya @_arnold_moya_]

@k8adev @femachini Haha. Passa o mesmo aqui em casa. Nossa bebê também quer codar.

[Arnold Moya @_arnold_moya_]

gramática inglesa actualizada 😆

[Arnold Moya @_arnold_moya_]

Smaller responses. Faster APIs. Less accidental data exposure. Cleaner contracts between backend and frontend.

A good DTO is not just a convenience.

It is a way to make the API speak with intention.

[Arnold Moya @_arnold_moya_]

One thing I’ve been appreciating more while building #API: DTOs are not just “objects to return data.”
They are boundaries.

[Arnold Moya @_arnold_moya_]

That’s where DTOs become powerful.

They let us return only the data that makes sense for the current context: the selected tenant, the user’s role, and the specific use case.