Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings
@asvarnon
asvarnon
Follow

Austin Varnon asvarnon

Block or report asvarnon

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
asvarnon /README.md

Austin Varnon

LinkedIn

I build tools to solve problems I'm facing, or because something seems interesting enough to apply a concept I'm working through — sometimes both.

Professional background in Java, Spring Boot, and enterprise backend systems at a large bank — containerization, OpenShift/K8s operations, CI/CD automation, and security remediation at compliance scale. Outside of work I run a two-node homelab for end-to-end project ownership: building, shipping, and operating software I actually use. Currently focused on Rust systems programming and local AI infrastructure.


📌 Featured Projects

Homelab MCP server

A Rust workspace that exposes homelab infrastructure (Proxmox, OPNsense) as AI-callable MCP tools — node/cluster status, DHCP leases, and more. Deployed behind Cloudflare Zero Trust with atomic CI/CD deploy and automatic rollback. Split into homelab-core (HTTP client, config, tool logic) and homelab-mcp (thin rmcp adapter).

Rust MCP Cloudflare

Self-hosted local LLM Discord bot

A persona-agnostic Discord bot backed by a local Ollama model. Threaded conversations, optional SearXNG web search, and two-tier memory — hot per-thread history in Redis, durable cross-thread recall via context-forge. Everything runs on your own infrastructure, no third-party API.

Rust Ollama Redis

Local-first memory library for LLM apps

A Rust crate providing persistent memory for LLM applications — SQLite + FTS5 BM25 retrieval, recency-decay scoring, and token-budget-aware context assembly. No network calls, no async runtime, no cloud dependency. Powers Husk's long-term memory.

Rust SQLite crates.io


🛠 Tech & Tools

Professional

Java Spring Boot Kafka Kubernetes OpenShift Docker Jenkins SQL PL/SQL

Homelab & Personal

Rust MCP Tokio llama.cpp cuda-oxide Proxmox OPNsense Cloudflare Redis Ollama SQLite


💡 Philosophy

I learn by building. Applying a concept until it becomes a skill is the only path that actually sticks. Most of what's here is that process made visible.


🔭 Currently

  • Rust systems programming — shipping production tooling in the homelab
  • Local LLM inference and AI-augmented infrastructure (Ollama, llama.cpp, MCP)
  • GPU compute fundamentals (cuda-oxide, Rust + CUDA)
  • Security tooling and compliance engineering (professional focus)

Pinned Loading

  1. connections connections Public

    Java

  2. mcp-homelab mcp-homelab Public

    Python 2

  3. EnvAndFoundationWorkshop EnvAndFoundationWorkshop Public

    Python

AltStyle によって変換されたページ (->オリジナル) /