1. X
  2. Donato Capitella
Log inSign up
Donato Capitella
2,568 posts
user avatar
Donato Capitella
@dcapitella
I'm a Software Engineer and Ethical Hacker, but mostly a tech enthusiast who likes to discover how things work by breaking them apart.
London
llm-chronicles.com
Joined September 2011
420
Following
1,094
Followers
RepliesRepliesMediaMedia
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

  • user avatar
    Donato Capitella
    @dcapitella
    Jul 17
    @UPS is the ONLY company that has not been able to deliver any parcels to my property for over one year, their refusal to engage in basic problem solving is tenacious.
    353
  • user avatar
    Donato Capitella
    @dcapitella
    Jul 14
    youtu.be/sMacAPpgXhQ I finally had a chance to play with AMD's new Ryzen AI Halo box, here I show my configuration that can get you 10-15% performance improvement on LLM inference.
    How I Configure the Ryzen AI Halo (Strix Halo) for 10-15% Faster Local Inference
    From youtube.com
    662
  • user avatar
    Donato Capitella
    @dcapitella
    Jul 7
    In this video how show how to build and run a Deep Research Agent from scratch using local AI (AMD R9700 or Strix Halo). I cover design choices, sub agent design and delegation strategies and how to benchmark.
    763
  • user avatar
    Donato Capitella
    @dcapitella
    Jul 3
    In this video I explore options to build a ervers for local AI using refurbished datacenter hardware (64GB VRAM : 4x NVIDIA V100/P100, AMD MI25)
    747
  • user avatar
    Donato Capitella
    @dcapitella
    Jun 25
    New Video: GLM 5.2 on Dual Strix Halo (256GB): Worth it?
    3K