Douglas Mendes

👨‍💻 Software Eng. Manager at @xpertsea ✝️ Christian 👪 Father and Husband. 📺 Neovim apprentice. 🧠 Grug Brained Developer’s

[CG @cgtwts]

many startups just died today

[Kol Tregaskes @koltregaskes]

Many developers have suspected for months that GPT-5.5 outperforms Claude Sonnet for coding. But SWE-Bench reported near-parity, and it made people question what they’d been seeing in practice.

DeepSWE aligns more closely with that day-to-day experience: GPT-5.5 scores 70% versus Claude Sonnet at 32%. That difference is substantial.

DeepSWE focuses on what tends to matter in real workflows: whether an agent can take a short behavioral prompt, locate the correct area of the codebase, and implement the change cleanly - without needing you to enumerate files, modules, and functions. SWE-Bench often fails to capture that, due to dataset contamination and weaker verification.
https://t.co/C3s80xfDkk

[Douglas Mendes @DougTheDev]

@theo I really want to see Apple merging or buying OpenAI. Very unlikely but cost nothing to dream ahhahahahahah!

[Douglas Mendes @DougTheDev]

@MrAhmadAwais @meabed @CommandCodeAI K2.6 is good choice as well to run on your CI

[Douglas Mendes @DougTheDev]

@meabed Bro, what Anthropic did is bullshit. I’ve been using claude -p to automate my harness tests, and now there’s suddenly a $200 cap on tool usage before they start billing separately through the API. That’s insane.

They are the new APPLE but for AI.

[Uncle Bob Martin @unclebobmartin]

I just cancelled my Claude account. I've been using codex, and haven't used Claude in several weeks.

[OpenAI @OpenAI]

Introducing Daybreak: frontier AI for cyber defenders.

Daybreak brings together the most capable OpenAI models, Codex, and our security partners to accelerate cyber defense and continuously secure software.

A step toward a future where security teams can move at the speed defense demands.

[Douglas Mendes @DougTheDev]

@elijahmanor @cpalions @accessurf Yes my brother!!!! ❤️❤️🙏🙏

[eric zakariasson @ericzakariasson]

this is cursor team kit: a plugin for some skills we use to build cursor at cursor

skills for verifying changes, driving local tools, and shipping reviewable PRs
https://t.co/8R4XNCUOfe

[OpenAI Developers @OpenAIDevs]

Pets. Now in Codex.

Use /pet to wake your pet.

[Douglas Mendes @DougTheDev]

@ancapchequei Mano do céu!

[Douglas Mendes @DougTheDev]

@trashh_dev I have been demoted from 'Father' to 'Automated Mining Unit in Minecraft.' My 6-year-old little dictator makes me find ores all playtime long, while she decorates the house and adopts puppies.

StarCraft would be heaven in comparison!

[Douglas Mendes @DougTheDev]

@theo @BoganBits I’ve stopped caring because people say it’s over every minute. They said that about GPT-2 and we’re still fine. I don't really trust what a CEO says for obvious reasons. Even if it is true and causes problems I think we will still be okay.

I love your content btw =)

[Douglas Mendes @DougTheDev]

After trying out a bunch of tools, skills, and the vanilla way of doing it, I’ve realized that, honestly, AI is pretty bad at writing tests.

[jack @jack]

https://t.co/jgZkBvYOPt

[Douglas Mendes @DougTheDev]

Bro! Huge!!!!!!

[dax @thdxr]

james has achieved distributed opencode

agents can run on your laptop, on a remote server, in a cloud sandbox provider

shut your laptop and things keep running

open it back up and all the data syncs

delete the sandbox nothing is lost

[Douglas Mendes @DougTheDev]

@adahstwt Well, not that hard to find:

[Douglas Mendes @DougTheDev]

@mitchellh @badlogicgames Little confession: I use ghostty with tumx

[Douglas Mendes @DougTheDev]

@Folke 100%