Daniel Vaughan

Co-founder at Orpius. Author. ex-Microsoft Technology and Research Engineer. Into Gen AI & Quantum. 🇦🇺 expat in 🇨🇭.

[Daniel Vaughan @DBVaughan]

I found myself repeatedly copying code out of Visual Studio to share it elsewhere.
Opening individual files, copying pasting the content of each file and wrapping everything in markdown.
So I put together a tiny extension that does it in one go.

https://t.co/RdJJBnLZts

[ASP.NET Core News @aspnetcore_news]

This week's ASP .NET Core newsletter is out! This week by @dotnet, @dbvaughan, @alihamzaansar16, @assiszang, and more ❤️ Check out 'Past newsletters' here: https://t.co/bvz2c3M16B

[Daniel Vaughan @DBVaughan]

My AI Slop Generator just notified me again. Some new slop at https://t.co/LtSDhK2Ugr

How I built it: https://t.co/Rk1LgdUL9X

[ASP.NET Core News @aspnetcore_news]

Building an Agentic Quantum Laboratory with ASP .NET Core by @dbvaughan https://t.co/ntEojPaT84 #aspnetcore

[Daniel Vaughan @DBVaughan]

Orpius v0.1.16 is out.
Now with out-of-the-box support for Moonshot's Kimi K2.5 model.
Use it immediately for scheduled, event-driven, and long-running agent work.

https://t.co/pfqZaWsKmK

#Orpius #AgenticAI #Kimi /cc @KimiProduct

[Daniel Vaughan @DBVaughan]

https://t.co/Rk1LgdUL9X

[Daniel Vaughan @DBVaughan]

I stopped competing with the internet and built an AI Slop Generator instead.

It writes, publishes, and shares posts on a schedule.
Agents, GitHub, workflows, secrets.

This is probably the future of content.
Agree or not?

[Daniel Vaughan @DBVaughan]

I've published a new article: Building an Agentic Quantum Laboratory with Orpius
It shows how to use AI agents to design OpenQASM experiments, run them via Qiskit and an https://t.co/S7YShNFQrU Core tool host. #quantum #AgenticAI
https://t.co/fYYNrcJIrD

[Daniel Vaughan @DBVaughan]

I miss the old solid yellow folder icons in VS 2022. The new hollow icons in 2026 reduce clarity for me. There's an issue for this over on the community site. Please up-vote if you're affected. https://t.co/9fhQUA0czm

[Daniel Vaughan @DBVaughan]

Come work with us!

[@orpiusAI @orpiusai]

We've just posted a new role at Outcoder.

This is a partner-style role, suited to experienced engineers, consultants, or solution architects who want to build recurring work around AI-enabled solutions.
https://t.co/vp5v1uEAuj

[Daniel Vaughan @DBVaughan]

Orpius agents write C# code, compile it, fix their own build errors, and run the result inside a locked-down WebAssembly sandbox with isolated storage.

This is coming client-side too, opening up local dev capabilities.

Have a play: https://t.co/Aq8CLot4CB

[@orpiusAI @orpiusai]

You can now try Orpius in a public trial instance.

Built in Switzerland and designed so your data stays inside your workspace.

Bring your own model and explore freely.

https://t.co/Pwj8W7vsnU

[Katka Vaughan @KatkaVaughan]

There is a public trial you can try right now: https://t.co/r8b2loqLaK

[Daniel Vaughan @DBVaughan]

@mitri_drake @libsoftiktok Oh, thank you for a sane reply. Couldn't agree more.

[Daniel Vaughan @DBVaughan]

Hugely excited our paper on Distributed Agent Reasoning is now on arXiv!
We go beyond centralized agentic systems and tool use, with agents on independent systems, overseeing their own private data, communicating across organizational boundaries.

https://t.co/O2kaPCvwu5

[Daniel Vaughan @DBVaughan]

Just used @windirstat to find large files in my docker context. I think I'm still using a windirstat version from the 2000s. It's one of those tools that is utterly indispensable when the need arises.

[Daniel Vaughan @DBVaughan]

Seems like 04-mini-high got a speed up today, but with a reduction in ability. @OpenAI, am I just imaging things?

[Kenneth Stanley @kenneth0stanley]

I think my former OpenAI colleague @nickcammarata makes a good point here that is very aligned with arguments from open-endedness, though said in his own unique way. I’d add that I think Nick is largely pointing to the perennial benchmark obsession that dominates the field. And this obsession is very related to the objective paradox we describe in Why Greatness Cannot Be Planned, which is the problem that the more you fixate on measurable progress towards a goal, the more likely you are heading into a brick wall and the less able you are to innovate. We should remember that the reason ChatGPT was one of the most quickly adopted technologies in history was not because people were wowed by some benchmark results, but rather because the qualitative experience was interesting, even fascinating! So for those itching to break out of the objective paradox, instead of thinking about how to ace a test (as tempting as that is), try to imagine what experience would be similarly fascinating to the general public without needing any benchmark result at all. That is the path of the interesting instead of the path of the objective, and indeed we often collapse to something like only 1% of effort given to the path of the interesting. Those 99% who are trapped wallowing in false security blanket of the objective paradox dwell there at their peril.

[Daniel Vaughan @DBVaughan]

I've noticed that Dall-e inside ChatGPT, uses the photoshop checkered background when you specify a transparent background. I wonder if this happens elsewhere.

[Hassan Habib @HassanRezkHabib]

Did you know that Microsoft Visual Studio will show you errors in your JSON string if you just put a comment //lang=json?

#json #csharp #dotnet