I never expected Docker to handle document management and AI chatbots, but here we are

Released Oct 11, 2026, 7:00 AM EDT

Previous radio and television guy(I Heart Radio and PBS). Now an author, scientist, and artist who blogs about the absurdity of our modern-day world and other unique subjects. Focuses on strange music, the old web, retro computer game and tech, and investigating weird, yet intriguing things. His work has actually appeared in Vice, Tedium, Star Trek.com, Today I Found Out, Typebar, SyFy Wire, and How To Geek.

Presently dealing with a range of music, web, and composing jobs. Visit him on the internet at https://djbuckfreelance.neocities.org/

With time, I’ve found numerous Docker containers/apps I can’t work without. FOSS apps are regularly premium, however some apps kick it up a notch. I liked the app a lot and got to thinking of comparable apps. Apps you may jokingly state “do things you wouldn’t expect from a Docker container” due to the fact that they’re premium-level, feature-packed programs that are (normally) totally free, self-hostable, and containerized. Here are a few of the very best ones I’ve discovered.

Paperless-ngx

For developing a searchable file database

We’ll begin with something near and dear to my work: file management. Paperless-ngx can deal with anything I toss its method. It lines up with my interest in structure and keeping a Knowledge Base in a smooth method while likewise permitting me to have a center for all my files.

Paperless-ngx is open-source and is indicated to develop a searchable database of physical files. The concept is you’ll scan or picture your files and publish them into the database. I’m utilizing it to handle expenditures, track music concepts and lyrics, in addition to developing a database of handbooks and chord charts.

Paperless-ngx utilizes OCR (optical character acknowledgment) to check out and classify the files you publish. To install it, I utilized the Docker Compose YML filehowever made a couple of modifications to get it to my preference.

As soon as I began the container with docker compose up -dI produced a login and began. The user interface is extremely basic and user-friendly. The concept is you simply drag and drop your files on the best side, and it’ll begin constructing the index from there. You can likewise alter OCR settings, AI settings, and include an application logo design.

For my preliminary test, I utilized it to arrange and support files associated to an album job I’ve been subjugating for the previous couple of years.

I set my OCR output to PDF and kept whatever in the basic setup. I made certain all AI functions were turned off, because I didn’t wish to utilize them for this specific task. There are going to be some exclusive files in here that I do not want to show LLMs. I included my logo design (a flying cheeseburger called Clyde) and began submitting files.

Establishing information for the file is quite simple. When you click the file or image, it opens a sub-menu where you can include the suitable information( identification number, developer, and so on )and tag it. This is an example of my “logo:”

From there, I started publishing suitable files: the tune list, lyrics, HTML files, chord charts/images, and other information pertinent to the album job. As soon as whatever was packed, I had the ability to

I understand, I understand. I’ve made it clear that I’m not a fan of LLMs in the past (mainly due to the fact that of the ecological effect, dodgy information, ridicule for AI art, and the method the weirdos in charge of it market them). That hasn’t altered. I do[ have an open mind about tech and throughout my research study, I could not miss the chance to speak about utilizing Docker to self-host a regional LLM.

I’ve really wondered about Ollama for a while now, so I took this as an indication to set up and take it for a trial run. It utilizes Open WebUI

Ollama has actually been offered as a Docker image considering that 2023I attempted it due to the fact that I like to self-host things and due to the fact that it can run offline without sending my information anywhere else. As soon as set up, I chose llama 3.2:1 b (I prepare to attempt 3b and mistral )as the design to run, considering that it’s quite light-weight and my test device might manage it.

I established the Docker Compose file and after that pulled the design straight in with docker exec -it ollama ollama run llama3.2:1b

The user interface is quite uncomplicated. The AI’s a bit sluggish, however that’s to be anticipated. I began by asking it about its constraints and abilities. I made it produce puns and a brain-teaser. It wasn’t excellent at puns, I’m sorry to state. It turned to daddy jokes quite quickly, however I did chuckle at the coffee cup being robbedPeak LLM humor.

The brain-teaser was a terrible failure. I addressed it, the design got puzzled and begun coding … something.

Let us never ever mention it once again.

I never ever would’ve anticipated Docker to manage an AI chatbot, however it works, and I can see a lot of possible usages in the future.

Docker continues to impress and amaze me

Often you discover an excellent app in an unforeseen location. Docker has plenty of those kinds of surprises. While I’ve just checked out a couple of here, I’m still finding brand-new and helpful software application. I’ve discovered Docker to be an effective buddy to my work, pastimes, and the simply plain ridiculous tasks that I like to do in my leisure time.

Read More