Hi!

I’m Michal. Recently, I built a Docker image for people interested in Stable Diffusion, especially LoRA training. I’ve spent thousands of dollars and countless late-night hours tinkering so you don’t have to. It took me more than a year to create a solid workflow for SD training, and along the way I realized I’d built a few genuinely useful tools. I started putting them on GitHub, then had another realization: why not publish the entire workflow and stack?

So I built a Stable Diffusion Docker image for RunPod (with Modal, Vultr, and DigitalOcean support coming soon), plus supporting tools, so you don’t have to wrestle with dependencies, CUDA mood swings, or “it worked yesterday” mysteries.

I’m an idea guy and was never great at coding. Then Stable Diffusion happened. Then vibe coding happened. I started working on my web projects and somehow ended up learning Python and JavaScript at 44 🙈

My projects are open, practical, and fueled by curiosity and a suspicious amount of coffee. If you find the tools useful and want to support continued updates, experiments, and bad ideas that somehow work, you can drop a donation here.

No paywalls. No guilt. No promises of world domination. Just better tools, fewer headaches, and more time generating cool stuff.

Thanks for being here!

Learn more about Lora Pilot at https://www.lorapilot.com