Skip to content
View AIandI0x1's full-sized avatar
💭
one step at a time
💭
one step at a time

Block or report AIandI0x1

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AIandI0x1/README.md

0x1

**Data Analytics · AI Governance · Agent Orchestration · Local/Cloud AI

Building and using 0x1 — Orchestra Unified, an AI-governed work orchestration platform that manages AI agents (1-10 cloud agents verified + local agents), verifies their output, and maintains a tamper-evident audit trail. 6,400+ Python modules, 14,310 commits, ~299K chained events.

My daily development driver. Contributions or any dev goes through my own System.

Where 0x1 sits in the layers

-Model Provider (API)- -Agent-Harness (hermes-agent, openclaw) ; IDE (soon called agent runtime) like Claude Code, Cursor, Codex, Devin et al.- -0x1 uses the above and others to achieve the operators Vision, goal and tasks- -OS (Mac, Linux et al.)- -Hardware-

Modular. Only take what you need. Made for continuous self improvement, loves hermes-agent wink wink Being fully your own Operator and developer (fully customizable) OS might be coming next year depending on recources Extensible

  • Orchestrating AI work across cloud and local — I am working hard to migrate 50-90% of all wort to the local execution layer which is bootstrapping on AMD Strix Halo (128 GB unified memory)
  • Multi-node setup: macOS + Linux over private mesh + Nvidia GPU Goal: multi node, hardware independent mesh networks able to utilize and manage recource pools efficiently and productiveley
  • Source code is private — showcase here

Main project

| Project | What it does

| 0x1-showcase | AI-governed work orchestration platform. Includes absorbed modules: agent harness (hermes-agent), communication platform (buzz), a couple of inference servers i will soon start to list em all up shout out to Carlo

| 0x1(https://github.com/AIandI0x1/0x1-Main-mirror) | The actual project early dev (coding) started two years ago, real full dev began September 2025, 0x1 as a vision and idea is 10+ years old now with AI i can finally accelerate development This month we will update to the next Version with major changes (August 2026), multi agent orchestration framework fully auditable, early alpha, Agent GOV; Authority, Synthetic Brain framework. State machine. Next steps: Fully Synthetic Brain APP (Dont fear the brain naming these are labels to boost coherence but can be chosen by the dev at will), Goal: Semi-autonomous AI factory driven by human in the loop. Continuoous self-improvement. Goal: Deploy a humanoid robot driven by 0x1 (preferably Unitree) EOY 2027

Hardware porting — AMD Strix Halo (gfx1151, RDNA 3.5)

Ported inference optimization tools to Strix Halo with documented evidence:

Project What was done
Hyperloom GPU type detection for gfx1151, llama.cpp framework adapter, 135 tests passing
Magpie Benchmark scripts for Strix Halo, TraceLens validation, hardware-agnostic trace analysis
GEAK RDNA 3.5 hardware knowledge, Triton and HIP kernel verification, wave32/wave64 guidance

All three verified on hardware: _autodetect_gpu_type() = 'strix-halo'

  • One step at a time.

Pinned Loading

  1. NousResearch/hermes-agent NousResearch/hermes-agent Public

    The agent that grows with you

    Python 224k 43.4k

  2. buzz buzz Public

    Forked from block/buzz

    A hive mind communication platform

    Rust