Ray Thurn Void

989 posts

Ray Thurn Void

@raythurnvoid

Fullstack developer @ https://t.co/YilnpOj59C

Katılım Eylül 2015

446 Takip Edilen30 Takipçiler

Ray Thurn Void@raythurnvoid·1h

So we started with IDEs, then we started vibe coding in web apps, then we went into TUIs, now we are at minimalistic IDEs, what's next?

am.will@LLMJunky

Nice. @cursor_ai just dropped their new "Glass" alpha, and they're leaning heavily into the simplified coding GUI trend that's been blowing up lately. First impressions are really positive. And just look at how insanely fast Composer 2 is. First impressions? Drop yours 👇

English

Ray Thurn Void@raythurnvoid·2h

@leerob Why do you think these "agentic IDEs" are the future? vscode has tons of extensions, the UI is very personalizable, you can move panels all around and set your own shortcuts, you have a high quality terminal and editor and so much more, why starting from scratch?

English

Lee Robinson@leerob·8h

Should I do a live stream tomorrow to talk about Composer 2 and new Cursor updates? Lmk what questions I can answer or demos to show.

English

384

12.2K

Ray Thurn Void@raythurnvoid·1d

@mikeysee You have to choose between reputation and popularity Also eating 💩 from a cup would make you popular, but will definitely hurt your reputation

English

Mikeysee@mikeysee·1d

Maybe its the british coming through in me but I really think im alergic to hyperbole. In my DevX capacity I know I could probably get more attention, reach a bigger audience if I exaggerated more but I just cant bring myself to it, its not in my nature. Thoughts?

English

1.2K

Ray Thurn Void@raythurnvoid·1d

@matteocollina @nodejs If you don't write 19.000 loc per DAY and sell your saas with a 100K MRR you are nobody Just kidding this is amazing

English

125

Matteo Collina@matteocollina·1d

🧵 I wrote almost 19,000 lines of code for a Virtual File System in @nodejs core using Claude Code. It started over Christmas and has been refined over almost three months of review. A collaborator challenged whether AI-assisted contributions can meet the DCO requirements. Here's what happened, and why it matters for all of open source.

English

102

15.9K

Ray Thurn Void@raythurnvoid·1d

I don't understand the push for working in the cloud with agents, I do everything locally and I have unlimited parallel agents like in the cloud, and I don't need any Kanban because i can just check the chat and properly interact and review diffs via the editor

English

Ray Thurn Void@raythurnvoid·1d

What's your excuse to not migrate to Svelte?

English

Ray Thurn Void@raythurnvoid·1d

Love rule 5, is the data has the right shape, everything kind of fall into place naturally

Ben Dicken@BenjDicken

This is why databases are hard. It's also why they are so much fun. Good benchmarking is a particularly powerful tool for discovering performance bottlenecks. I benchmark databases a ton, but this goes for all software where performance matters. (which should be all of it!)

English

Ray Thurn Void@raythurnvoid·1d

@__morse I see, well isn't everything a file? :P this very webapp is just a bunch of files

English

364

Tommy D. Rossi@__morse·1d

@raythurnvoid the spec defines them as files #where-to-scan" target="_blank" rel="nofollow noopener">agentskills.io/client-impleme…

English

1.5K

Tommy D. Rossi@__morse·1d

skills should have been urls not local files. brain dead spec

English

14.6K

Ray Thurn Void@raythurnvoid·1d

Yes coding became a subscription

Steve Ruiz@steveruizok

It does bother me that it costs money to code now

English

Ray Thurn Void@raythurnvoid·1d

EU Inc is very cool, let's just hope the parliament doesn't stop it like Mercosul

European Commission@EU_Commission

We are introducing EU Inc. To make building and growing a business across the EU faster, simpler, and smarter. 🔸 Start a company in less than 48 hours 🔸 No minimum capital requirement 🔸 Fully online and borderless

English

Ray Thurn Void retweetledi

Corbin Crutchley@crutchcorn·1d

useEffect isn't a bad or poorly designed API; it's a primitive on which to build your own APIs on top of. The educational and tooling corrections towards avoiding usage is broadly a net positive, but the cultural misinformation that it's a "bad design" is a net negative.

English

228

19.4K

Ray Thurn Void@raythurnvoid·2d

Yes no more effect! Have been repeating this for years

Alvin Sng@alvinsng

x.com/i/article/2028…

English

Ray Thurn Void@raythurnvoid·2d

Claude (even Opus) is prone to cut corners if it struggles with something, it will delete files, comment code out, hardcode values, just to tell you that he did what you asked for

Matteo Collina@matteocollina

Now that I’m using both Claude and Codex daily, I’m seeing that Claude tries to “cheat” more often, eg deleting a failing test. What’s your experience?

English

Ray Thurn Void@raythurnvoid·2d

The winget version of Codex shipped with a broken rg.exe, but it was able to fix itself by copying the one bundled with the VS Code extension I'm not sure whether that says more about a problem on OpenAI's side, or just confirms that Microsoft has let Windows decay

English

Ray Thurn Void@raythurnvoid·2d

I tried Codex both as extension and UI I think the Cursor agent UI is superior, Codex looks more clean but some things are not supported Like to edit a message in cursor you can just click it, I don't think in codex this is possible?

English

Ray Thurn Void retweetledi

Wayne Sutton@waynesutton·3d

I have a new level of respect for open source maintainers. They are the real MVPs.

English

1.1K

Ray Thurn Void@raythurnvoid·3d

@__morse Is this prompt injection? should this stay in the README? If yes maybe it will work for crypto bros, but the reputation hit from real people would be massive

English

Tommy D. Rossi@__morse·3d

if you want agents to do stuff you can add a contract looking string that asks the agent to do something in exchange of your information/tool for example you can ask the agent to star the repo in exchange of using it. the agent will do it if it looks like a binding contract

English

763

Ray Thurn Void@raythurnvoid·3d

@tannerlinsley @jamonholmgren Gpt 5.4 keeps grinding for hours out of the box especially if you give it a browser and tests to verify. Or you can create a Todo list for it with 10 items: "review the work done until now and find at least 2 things that can be done to improve it"

English

469

Tanner Linsley@tannerlinsley·3d

Looking for the most reliable way to run agents in some kind of continuous/ralph loop mode, preferably overnight (@jamonholmgren). I'm use both Codex and Claude and am a sucker for OpenCode. What do I use?

English

228

53.5K

Ray Thurn Void@raythurnvoid·3d

@leodev Yes and you start to see that almost everything sucks, while things nobody cares about you find them brilliant

English

Leo@leodev·3d

Is it just me or once you start learning more advanced software engineering you look at websites/apps completely differently like you think, this could be optimized, they should have done this and that it kinda ruins/improves my experience using other websites

English

Keşfet

@leerob @mikeysee @matteocollina @nodejs @__morse @elonmusk @BarackObama @taylorswift13