Joseph Kaiser

Small tools, self-hosted systems, and model experiments.

I build lightweight software that is useful, easy to run on my own hardware, and pleasant to maintain.

Most of my time goes into practical web apps, Raspberry Pi friendly services, and LLM evaluation work.

What I keep building.

The work that keeps pulling me back tends to live at the intersection of useful tools, personal infrastructure, and applied model research.

01

Self-hosted tools

Small applications that run comfortably on home hardware and stay easy to update, debug, and maintain.

02

Model evaluation

Fine-tuning and benchmark experiments aimed at understanding where models improve and where the baselines still fall short.

03

Lean interfaces

Web frontends that surface just enough information to be useful, without adding operational weight or visual clutter.

What I build with.

A tighter look at the stack around the work: the languages I reach for, the runtime pieces that keep things light, and the hardware I like deploying on.

01

Languages

My default toolkit stays close to durable, low-overhead languages that make automation, UI work, and backend experiments easy to keep shipping.

  • Bash
  • HTML/CSS
  • Python
  • TypeScript
  • SQL

02

Frameworks & runtimes

I prefer compact runtime choices that stay understandable in production: containerized services, SQLite-backed apps, local model serving, and small web frontends.

  • Docker
  • SQLite
  • vLLM
  • Static sites
  • LAN services

03

Hardware & deployment

Most of the fun comes from owning the full stack end to end, especially when the software can run comfortably on small machines and stay easy to maintain at home.

  • Raspberry Pi
  • Home lab
  • Self-hosted Linux
  • Private network
  • Local-first ops

Selected work.

A short list of projects that represent the kind of software I like to make and maintain.

01

ai-chat

A web UI for a locally hosted Dockerized vLLM model, built for private LLM access over a LAN.

  • Docker
  • vLLM
  • LAN
View project

02

myrssfeed

A self-hosted RSS reader for a Raspberry Pi, designed to stay fast, simple, and easy to maintain.

  • Raspberry Pi
  • RSS
  • Self-hosted
View project

03

Current focus

Fine-tuning and evaluation work aimed at pushing a model past baseline benchmark results.

  • Benchmarks
  • Fine-tuning
  • In progress

In progress

Let's connect.

If you are building something useful in web tooling, self-hosted infrastructure, or model systems, I am always happy to compare notes.