Skip to content
View austinp0573's full-sized avatar

Block or report austinp0573

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
austinp0573/README.md

Austin Porter

I build software for systems, infrastructure, and backend—Linux, networking, and cloud. Most of what I publish here sits in that space, from homelab automation to edge deployments.

What I work on

Day-to-day interests: platform-style engineering, cloud and on-prem infrastructure, and networking. I run a homelab built around Proxmox, pfSense, and self-hosted services as a place to test IaC, upgrades, and service design before they touch production-shaped problems.

Technical skills

Languages: Python, IaC, Bash, SQL, Java, C#/.NET, C/C++

Stack & tools

  • Multi-cloud: AWS/Azure/GCP
  • IaC & automation: ansible, terraform/opentofu, jenkins
  • Containers & image supply chain: Docker, Kubernetes, Packer
  • Systems & network infrastructure: Linux administration; switching, routing, firewalls, wireless
  • Delivery & source control: Git
  • Backend & services: Spring Boot, .NET
  • Computer vision & embedded / hardware: YOLO / computer vision; Raspberry Pi, NVIDIA Jetson, microcontrollers, PCB design

Selected repositories

Real-time person detection on NVIDIA Jetson Orin Nano Super using YOLO11, with a C# / .NET 8 service and automated email alerts.

Terraform provisions Kubernetes node VMs on Proxmox; output feeds Ansible inventory for cluster configuration.

FastAPI app deployed on Heroku. Live instance: tusko.org.

The original iteration of my homelab worked, but it was all ad hoc. It felt dirty not having everything properly reproducable and documented, so I'm starting from scratch doing it the right way. With as few exceptions as possible, it doesn't happen if it's not in code, and tracked with git.

Time-series / ML experiment in Python and Jupyter: closing-price prediction from historical market data.

Install script for proprietary NVIDIA drivers on Debian 13; maintained for community use (currently targeting 580 series for stability).

Custom Quectel LC86G-PA breakout: schematic, layout, and bring-up for GNSS prototyping.

Focus areas

  • Self-hosting and infrastructure automation
  • Kubernetes operations and HA patterns
  • Cloud architecture and services
  • Advanced networking (pfSense, enterprise-style routing)

Contact

linkedIn · email

 

466f724a616e6574

Pinned Loading

  1. nvda_close_predictor nvda_close_predictor Public

    Use machine learning to predict the closing price of NVDA at a user defined number of days in the future

    Jupyter Notebook

  2. outside_probably_sucks outside_probably_sucks Public

    Call a weatherAPI and have it send you an email depending upon... whatever you want.

    Python

  3. raspberry_pi_pico_2w_various_modules raspberry_pi_pico_2w_various_modules Public

    Scripts and libraries for quickly getting started with the Raspberry Pi Pico 2 W and various sensors/modules. Each folder contains code for a specific sensor or module, including examples and libra…

    Python 1

  4. yolo11_jetson_orin_nano_super yolo11_jetson_orin_nano_super Public

    Getting ultralytics YOLO11 natively installed and running on a Jetson Orin Nano Super

    Python 1

  5. Hardware Hardware Public

  6. debian13_nvidia_drivers debian13_nvidia_drivers Public

    script to get nvidia proprietary drivers on a fresh debian13 install

    Shell 2