Marcelo

Software Engineer(15yoe)
Creator of

[Marcelo @itsmemarcelorm]

I kept running one coding agent at a time and sending every step to my priciest model. The boilerplate, the tests, the planning, all of it billed like hard reasoning.

So I built AOP. You set up the pipeline once (implement, test, review, debug, security) and pick which model and how much effort each step gets. A cheap fast model handles the scaffolding, and the heavy thinker only shows up where it earns its keep. I'm spending a fraction of the tokens now.

It also runs your own skills. Your prompts and methodologies become reusable steps, and they live in your repo with the rest of your code instead of some global skills folder you forget about.

You can mix Claude Code, Codex, opencode and pi in the same pipeline, with retries and completion signals so a worker knows when it's actually done.

It all runs on your machine, with a local SQLite file and no account to create.

Four workers free:
curl -fsSL https://t.co/a9kHyRzZiZ | sh
Still alpha, I'm shipping fixes most days.

[Marcelo @itsmemarcelorm]

@haider1 how many TPS are you getting with ZAI Pro SUB ?

[Marcelo @itsmemarcelorm]

@zachlloydtweets I created a tool that helps you to manage multiple models/effort/fastmode. Take a look, it's free!!

https://t.co/X9yKHT12Oj

[Marcelo @itsmemarcelorm]

@rauchg Imagine using with my product https://t.co/X9yKHT12Oj and be able to create workflows using multiple skills and models.

[Marcelo @itsmemarcelorm]

@MatthewBerman Download my tool! https://t.co/SoLmnMWHn2

[Marcelo @itsmemarcelorm]

@mweinbach Do you guys know if the problem is on the Model or on Codex? I've been using with pi.

[Marcelo @itsmemarcelorm]

@AkitaOnRails Consegue colocar o GPT 5.4 - High ?

[Marcelo @itsmemarcelorm]

@narek_gevorgyan @theo my pleasure!

[Marcelo @itsmemarcelorm]

@theo @narek_gevorgyan @jullerino Glad to help ! <3

[Marcelo @itsmemarcelorm]

@theo @narek_gevorgyan He is running under Rosetta, @narek_gevorgyan you should download the arm64.dmg on github releases !

[Marcelo @itsmemarcelorm]

@theo @narek_gevorgyan It's not a bug, I sent you a dm @theo , this is because the default dmg download is x64 not arm64

[Marcelo @itsmemarcelorm]

@Archetapp @martinlasek There is no comparisson, those chinese models were all benchmaxed, if you use daily as an engineer you will notice that is all crap compared to Claude Opus or Codex.

[Marcelo @itsmemarcelorm]

@bcherny @bcherny Can you add bypass permission on claude code desktop ? I wanna switch from CLI, but without that is impossible.

[Marcelo @itsmemarcelorm]

@skirano where are you using ? Cursor ? Codex? Is there any difference in results from your perspective ?

[Marcelo @itsmemarcelorm]

@antigravity When will Ultra subscribers get higher limits?

[Marcelo @itsmemarcelorm]

@apples_jimmy Meanwhile google gave us unlimited access to 2.5 pro on release.

[Marcelo @itsmemarcelorm]

@github @OpenAI @ashtom I'm a PRO+ user, any idea when upload images/screenshots will be available on vscode copilot for o3 and o4-mini?