Mariano Giusti

Software Engineer working at Red Hat. Self hosting and AI agents enthusiast, with a background in aerospace engineering.

[Mariano Giusti @marianoilconte]

@imbabybrooklyn I think it has actually! Not measured in any way, just my perception

[Mariano Giusti @marianoilconte]

@MiaAI_lab Ready to test this, awesome!

[Mariano Giusti @marianoilconte]

@TechHutTV @Burito_Joe @SynorosHQ @tenobrus I’m just too invested in Tailscale for my homelab at this point, but Netbird is great yeah

[Mariano Giusti @marianoilconte]

@Burito_Joe @SynorosHQ @tenobrus I use netbird at work

[Mariano Giusti @marianoilconte]

@SynorosHQ @tenobrus I’d like to do that too, I want to install Tailscale and proper Android apps on my Samsung TV.
Same thing with my older 4K Firestick, it’s stuck with an old android version and I can’t install Tailscale on it, it would be the perfect portable solution to watch my Jellyfin stuff

[Mariano Giusti @marianoilconte]

@DomAtSiteSage I finished my weekly limits 4 days early but I think next week I’ll build my Hermes bot app for pebble, if I manage to finish the other 3 projects in time, and also I’ll have to play with the Sparks

[Mariano Giusti @marianoilconte]

@SynorosHQ @tenobrus tell me more about it lol

[Ash Hart @ashxhart]

Every open model ships frozen.

What if it could learn and update its own weights as it's being used?

I have been working on a bit of engineering called Living Weights for about three months, and I think it's time to release it.

The model becomes uniquely yours, no KV stashing, no MD files, no trickery.... just live learning. 😎

Unload the model, move it to another machine, and it still knows what you told it. Could this be the start of recursive self improvement for local AI?

It drops in the latest TensorFold release today, with support for @NVIDIAAI Nemotron Lightning. 🤯👀

@abheer @sundeep @NaderLikeLadder

Special thanks to @volatilemarkts for the name.

[Alex Ziskind @digitalix]

Trying to bring up Omarchy on the Snapdragon X2 Elite mini PC. building device trees and all that mess.

[Mariano Giusti @marianoilconte]

@Teknium Next week I’m gonna cook something with my Pebble watch and ring using the gadget SDK and Pebble’s C SDK, I feel so inspired

[Joshua Warren @JoshuaSWarren]

Running AI on a Mac's GPU under Linux was painfully slow.

So I went into Mesa, the open source GPU driver, and made it 18x faster.

Decode: 1.5 to 27.4 tokens/sec.
Prompts: 55 to 237 tokens/sec.

On an M1 Max with Omarchy Linux.

It's all open source: https://t.co/7cVpy3mAaJ

[Mariano Giusti @marianoilconte]

@Teknium Applies 100% to me as well, you reignited a passion for SWE that I was kinda losing at my current job. I’d love to work at NousResearch one day, it must be hella fun. I like to work doing things that are meaningful to me, open source harnesses are all that matters to me now

[Mariano Giusti @marianoilconte]

@thsottiaux Can we make it possible to see and even activate banked resets from the mobile app?

[Mariano Giusti @marianoilconte]

@ashxhart I’d love to help with Tensorfold now that I’ll set them up

[Mariano Giusti @marianoilconte]

@HermesWatcher I lost track

[Mariano Giusti @marianoilconte]

@Teknium And that’s why I love Hermes

[Mariano Giusti @marianoilconte]

@tonysimons_ Opus 5.5 hands down

[Nous Research @NousResearch]

Introducing Hermes Index: an average of benchmarks measuring model performance in Hermes Agent.

Our goal with Hermes Index is to better inform users choosing between models and to show labs how they perform in Hermes.

https://t.co/GMkIS99lcT