Skip to content
View XXY-CH's full-sized avatar
:bowtie:
:bowtie:
  • Hangzhou XueJun High School
  • Hangzhou
  • 05:59 (UTC +08:00)

Highlights

  • Pro

Block or report XXY-CH

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
XXY-CH/README.md
CachoidXie


root@kernel:~$ cat /proc/identity

I build research-shaped systems: agent harnesses with auditable authority boundaries, online judges that understand teaching data, and memory architectures that try to make long-context reasoning cheaper, sharper, and more falsifiable.

My favorite work lives where a mathematical claim has to become a running system — proofs, tests, architecture diagrams, telemetry, and a repo another engineer can clone, compile, and reproduce.


⚡ Active Systems

Local-first Agent Harness Kernel

An auditable runtime for agent execution, memory, permissions, capabilities, and governed tool use. Git-like control plane between a user, their devices, their data, and autonomous agents.

AI-Native Online Judge Platform

Multi-tenant competitive programming platform for educational institutions. Secure sandboxed execution with cgroups + seccomp, real-time WebSocket updates, LLM-assisted analysis, and 6-level RBAC.

Proof-Aligned Long-Context Memory

PyTorch research scaffold decoupling static memory lookup from dynamic reasoning. RetNet recurrence, Block Attention Residuals, and hashed N-gram Engram lookup — backed by 42 formal proofs.


🗺️ System Map



Aetherion
Agent Harness Kernel
CodeNexus
Education & OJ Platform
Anamnesis
Long-Context Memory

🛠️ Tech Stack

Layer Technologies
Systems Rust C++ Go TypeScript Linux
AI / ML Python PyTorch Memory Models Agent Scaffolds
Infra PostgreSQL Redis Docker Git

📐 Engineering Axioms

"Simplicity is a prerequisite for reliability."Dijkstra

# Principle
1 Executable First — Start with the running system, not the slogan.
2 Abated & Proven — Keep claims narrow until tests make them stronger.
3 Falsifiable Design — Architectures should be inspectable and disprovable.
4 Evidence-Centric AI — Build on empirical evidence, not chat wrappers.
5 Docs as Code — Documentation is a core component, never an afterthought.

📊 Telemetry

Field Value
Host XXY-CH@github
Status active
Repositories 7 public
Focus Aetherion · CodeNexus · Anamnesis
Research doi:10.5281/zenodo.20041183

Pinned Loading

  1. NEW-Finding-Shortest-Paths-Algorithm NEW-Finding-Shortest-Paths-Algorithm Public

    A brand-new shortest path algorithm comes from the latest research of Tsinghua University.

    C++ 2

  2. CodeNexus CodeNexus Public

    CodeNexus is an online judge platform that developed for school to connect teaching with AI. The AI connecting gateway is still in processing

    Rust 2 1

  3. Anamnesis Anamnesis Public

    PyTorch research scaffold for RetNet, Block Attention Residuals, and hashed Engram memory

    Python 3

  4. Aetherion Aetherion Public

    Local-first Agent Harness Kernel with Git-like control for safe, auditable AI agents.

    TypeScript 4