I started playing (learning) guitar once again.

Now that AI is here it's much easier to approach. It won't help fix my finger positioning but it helps a lot with exploring the theory.

I built a library for visualizing and playing everything: https://dsebastien.github.io/music-notation/

Then an Obsidian plugin on top: https://dsebastien.github.io/music-notation/

And an LLM wiki about the theory: https://notes.dsebastien.net/10+Meta/99+AI+Assistant/Wikis/M...

AI slop certainly but it helps me a ton!

Very cool. I just started using AI to do something far simpler. I will definitely check out your site.

I had restarted guitar after a 45 yr absence of playing. Took lessons for 9 months and progressed well, but hit a wall. I was going to look toward AI for help getting over the wall what you did looks promising. Thanks!

I use a single Knowledge Base as the starting point of each and every AI session. That knowledge Base doubles down as an Obsidian vault in which I also have various databases, plugins, etc).

My knowledge base contains notes about my goals, projects, tasks, products, ideas, etc. It also contains my articles (dsebastien.net), my tools and infrastructure, my open source and private projects with their repositories, specific instructions for each, etc.

In addition, I have built tools to help understand the taxonomy/ontology of my knowledge base, know which types of notes exist, which properties, which template to use, where to store what, etc.

On the human side, everything is automated, which lets me capture efficiently. On the AI side, all the knowledge is easy to find and leverage.

The whole knowledge base is like composable context for AI.

I can start sessions with sentences like "let's work on the music practice plugin; add xxxx" and AI will understand what project it is, where to find the sources, where to clone or update the repo on my machine, etc. And from the context within the repository itself, it will then know how the code is organized etc. The nice thing is that this also enables me to work on multiple projects in parallel easily; either in different AI sessions, or in a single session for work that spans multiple projects/repositories. For instance I have a repo which a product, another with the product documentation. When I ask for something in the context of that project, AI works seamlessly on both.

BTW this product of mine is a 1:1 copy of what I use for myself: https://store.dsebastien.net/osk

I've always downloaded entire YT channels I enjoyed. I now have free/forever access to tons of content that now sits behind paywalls on topics I care about.

All automated for years

What are you using?

Not the poster you're replying to, but my archiving rig uses my personal fork of "tt-rss" to pull in the RSS feeds for the channels I care about and a script to query the database. That script uses yt-dlp to download the videos and subs, then marks them downloaded in the database. It's a bunch of glue-and-tape scripting code but was easy to hack together in an afternoon. Tweaking yt-dlp to run at a velocity that didn't get my IP address put into "you must login" hell was the most difficult part. The only long-term care-and-feeding has been keeping yt-dlp updated in the face of the arms race.

> Tweaking yt-dlp to run at a velocity that didn't get my IP address put into "you must login" hell was the most difficult part. YouTube is pretty hassle-free in that regard, at least for my IP region.

I have a script that downloads sequentially, and there’s an optional variable interval in between—though I don’t think I’ll actually need that. I once downloaded the entire contents of a YouTube channel that had over 400 GB of audiobooks. After a while, YouTube throttled the speed to 1/2 MB/s—which didn’t bother me, but everything still downloaded successfully. That was already two years ago, though.

A few days ago, I spent two consecutive days nonstop downloading playlists from YouTube Music, but that didn't trigger anything either.

I stick a random delay between downloads, never download streams in parallel, and always download when nobody is using YouTube (ideally when nobody is home or everybody is asleep).

I never did get throttled-- I just stuck in "you must login" hell. We don't login to YouTube in our household, so that effectively cut YouTube off for my wife and daughter. My wife couldn't have cared less but I learned that being able to stream music videos while getting ready for the day or bed is mission-critical for my daughter. (I've asked her to identify channels I can sic the archiving rig on but, heretofore, she hasn't done it. I keep telling her those videos she likes could disappear at a moment's notice...)

I created an AI skill for this a while ago: https://www.dsebastien.net/garmin-ai-skill/

I built ghradar.dsebastien.net recently because I needed something like that

Yes there definitely are.

I created a custom status line to see/track everything I need: https://github.com/dsebastien/claude-epic-status-line

That avoids surprises

Yes and they have a specific subscription for managing sandbox policies across the enterprise: Docker AI Governance

You can create those manually but if you want to enforce those then you need the subscription

The one thing I wonder about is how you enforce the usage of Docker Sandboxes vs running the agent on the host directly, apart from scanning machines for binaries

Add me there