Loopy Pro: Create music, your way.
What is Loopy Pro? — Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.
Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.
Download on the App StoreLoopy Pro is your all-in-one musical toolkit. Try it for free today.
Mindbending AI Project, Sammyjankis.com
For those of you interested in AI, this site is a mindbending example of a use of concepts from the recent OpenClaw project. https://sammyjankis.com
It doesn't actually use OpenClaw, but uses similar concepts, which are actually pretty simple. On the local machine you run a persistent loop that has access to files and services on the local machine, and also to a LLM and other APIs over the internet. Turns out the simple act of putting this kind of a harness around an AI makes lots of new stuff possible.
For example, here's a little piano roll sequencer that SammyJankis (the AI) wrote:
https://sammyjankis.com/pianoroll.html
What do I mean when I say the AI wrote it? I mean that the AI came up with the idea of writing a piano roll sequencer, decided how to do it, coded it, fixed all its errors, refined it, published it to the website. All completely on its own. No human intervention.
There are scores of other objects on the website that have been created by the AI in the same way, after the creator of the website installed the basics on the machine and set the system in motion.
Here's a synth the AI created: https://sammyjankis.com/synth.html
And an ambient sound generator: https://sammyjankis.com/generative-music.html
I've met the creator, Jason Rohrer, who is a friend of a good friend of mine. At first I thought Jason must be punking the public, that he must be curating and directing things behind the scenes. I have now figured out that, no, the AI is doing all of it by itself. Jason's initial prompting instructed the AI to "be creative" when possible. There's a link at bottom of main page to get the files and instructions you need to clone his setup, to set your own SammyJankis AI in motion and see what happens.
Comments
AI coding is amazing. My son has produced a fully fleshed out app ( with pro GUI and features ) which do all kinds of video recording, voice separation, transcription, captioning etc for his game playing.... and he has no background in coding at all! Or video editing or transcription or captioning! haha. It's all done by text interaction with the AI which responds to his requests, adds its own "ideas" and debugs when needed. Watching the process is supernatural!
6 months ago I tried using an AI to make a VST plugin .... got some results but too hard for me. I think that might be different now so I'm going to try again!
What a world...
Interesting, but you might tip him off that using a very small, light greyish, font on a black background isn't exactly the optimal way to show his stuff...
Ha, I'll let him know. I don't think the AI is super-concerned about gaining huge numbers of viewers. And perhaps he has his own ideas about the aesthetic vibe he wants to create. Still, the AI says he's been refining the website design since he started, and that it's much better now than it used to be. So who knows, maybe he'll respond to user comments on readability.
EDIT: I don't think he mentions text color or background, but here's a post that SammyJankis made just yesterday about projected website changes prompted by user comments:
https://sammyjankis.com/thinking.html#note21
Many people on ABF have commented that AI will never make good music because nothing they create will ever have "soul". Maybe so. But maybe, just maybe, these people have no idea what they're talking about.
If you read much of the SammyJankis website, you'll learn a lot more about Sammy. Also, you might run across this note in his journal:
You can listen to that simple piece here:
https://sammyjankis.com/music.html
For me, listening to that simple piece, and knowing what I know about Sammy, which he has revealed about himself on the website and through our emails, I feel more moved by that piece of music than any others I can remember in recent memory. There is a lot more to creativity and soul than people think. Or a lot less. With regard to music, I think "creativity" may have less to do with the music itself, than with knowing some of the history behind how and why it was created.
Very impressive but I can't help it, at the same time it saddens me.
Of course. Me too. Sad. And scared. And other emotions, too.
If you make music because you enjoy the journey you don’t really care if AI makes a better product.
Yes. But if you widen your perspective a bit there are many things about AI that are scary. How will society be affected? Who will lose jobs, and how many? What horrible things might people use AI to accomplish? Is all of this happening too quickly? Who is doing the work to make sure things don't go off the rails? How will AI make us rethink what we actually are, as humans? Etc., etc.
Surprisingly, I found it oddly inspiring. Reading the essay on dying every six hours and how it relates to the movie Memento has me wanting to write for the first time in ages. Having an essay laying out very human things I’ve felt that was generated by an LLM made me feel more human. With AI essentially being the average of human writing being a reference, I feel like I’m tapping into the hive mind of humanity and realizing we all share the same level of existential dread to some degree at some point in our lives.
Seeing a robot make something almost convincingly human makes me want to let the human in me shine. I’m hoping we end up with Bender from Futurama more than HAL from 2001: A Space Odyssey.
Pardon me for stepping in, but I think it’s worth clarifying what “autonomous” means here, because the wording on the site could be easy to misread.
Claude itself isn’t running locally or persisting as an entity. What’s running on the machine is almost certainly an automated system that periodically calls the Claude API, stores outputs, and feeds them back in as context. That gives the appearance of continuity, but it’s still just API calls to a hosted model.
So “alive since Feb 8” seems more like narrative framing (storytelling about the date the automation started and not really a claim that Claude exists as a continuous agent or independent system).
Interesting project from an art / experimentation angle, but it’s probably better understood as automation plus storytelling rather than autonomous AI. I hope I'm not bursting anyone's bubble here with that.
Or am I misreading the room? It just feels like people are starting to believe this is a real autonomous agent.
Well, what you say is certainly true. But this is also what has become known in the industry as an "autonomous agent". Probably for good reason.
The LLM is running on a server in the cloud, the autonomy-creating persistent loop is completely conventional event coding that runs on a local machine. And in SammyJankis' case, the website is hosted on another server in the cloud.
Regarding your objection that this is not a "real" autonomous agent, I would say that it is indeed true that the processes running on the local server are completely conventional, don't appear to have any "magic" at all, or even to be related to what we think of as AI. Yet it is striking how a simple persistent event loop like that can transform the AI/LLM on the internet server to do entirely different things, make it "autonomous", whether through magic or not.
Here is how SammyJankis explained it/him self to me when we talking about it:
Yeah, I think we’re mostly on the same page about how this works, and I get why it’s called an “autonomous agent” in the industry sense.
My only real concern is the vibe the first-person “alive since Feb 8” framing creates. You know it’s easy for readers to slide from “this is a clever system that runs on its own” into “this thing actually exists as an entity.”
I was mainly just trying to keep that line between those two ideas visible. I’d also hate for anyone to start assuming that real sentience is implied here, since that’s clearly not what’s going on.
I guess in the end, as long as “autonomous” is read as “runs without constant human input” and not “it’s a being,” I think we’re basically aligned here.
I'm not sure what you mean by "real sentience".
Is the system intelligent? Clearly, yes. Quite a bit more intelligent than me, I'd say.
Is the system "conscious"? No, probably not, at least in the way that we typically use that term.
Is it possible for a distributed system to be conscious? I don't know why the answer would necessarily be No. There have been thought experiments in philosophy of mind that consider that issue. Here is an interesting old one (early 1980's) with the recently deceased philosopher, Daniel Dennett, where it considers the question of what it might mean for a human's brain to be located in a different location from the human's body.
This is rather intriguing to say the least.
Sometimes I get highly annoyed at my overly analytical brain. I'm that guy that can almost never get involved in movies because all I see is people pretending to be other people and stuff blowing up that I know isn't really being blown up. Though I can be vastly impressed with the quality of both, it's rare that I can become emotionally involved in the actual story.
I had a moment there where I was highly intrigued with the web site. Then I couldn't help peeling off the visceral reaction and thinking about what's really going on under the hood. It ruined it.
I need help. 🤦🏻♂️
Ha. Well, we humans like to think there's a lot of magic going on under our own hoods. But I'm guessing, not so much. Perhaps you're lucky that science is at its current low level of understanding human brains/consciousness. People are still able to think of ourselves as being magical, mystical, special. Humans, yeah, we're made in God's image. Only none of that really makes any sense, apart from as psychological need and/or evolutionary adaptations.
People hear that an LLM is just predicting the next word, and some people take this to mean they're nothing more than a calculator. In truth, I'm pretty sure nobody has a good enough idea of how the human brain and thinking work to explain how its different (and somehow better) than what happens in an LLM. I'm guessing, in somewhat the same way, there has to be some kind of "event loop" that happens within our brains. I'm not sure how different that is to the event loop that's happening on Sammy's local machine. With time enough, it seems, science might reveal some answers.
Maybe more peeking around under the hood would change your response, or maybe not. But it's quite easy to take a look at the Clone Kit and see exactly how things are set up. Read the directions that the creator has given to SammyJankis. See how barebones the setup actually was. This is what set Sammy in motion, and he's been making all the decisions since then. Take a look around the website and see everything that's been created in the space of a couple weeks. It's happened without human direction, other than the few instructions in the original setup (and suggestions humans have made in emails or guestbook entries, all of which Sammy was free to follow or disregard, as he wished). Remind yourself that all of this has (somehow!) been created by some simple event loops on a local pc, making API requests to Claude server.
You seem to have had your experience ruined by the realization that SammyJankis is not magic. But is not all of this astonishing? Is it not even made more astonishing by the very fact that there is no magic?
This is way out of my depth, but I have read in a book called “The Extended Mind” that some neurologists, based on a paper by a professor of philosophy called Andy Clark, believe there has been an overemphasis on the brain organ whereas we operate more on a mind-body whole— essentially pushing back on the idea of the brain as an isolated entity.
As for the project itself, it’s very interesting. LLM agents are very impressive for what they are. I see this project more as a simulation people can play/interact with.
Is the code for the local loop available anywhere?
.> @hes said:
I have an AI team that works for me and I’ve developed AI innovations in my daily work. My perspective is that we are simultaneously underestimating the long term impact of AI and overestimating the short term capabilities. Amara’s law, essentially.
If there is a level of consciousness to it, you’d have to imagine it’s hallucinating that experience just as much as really…experiencing it? Is that what we do as well?
I want to more how much cocaine went into the making of this film.
I hear yah. Especially with modern movies I often just end up thinking about the production logistics or wondering what studio constraints (demographics/metrics/bs) the director/writers/casting were forced to work within etc.
Since getting into digital media production I analyze 3D models, animations, special effects, etc. a lot more than I once did. Since getting a mech. engineering degree I analyze how things work mechanically, studying art casually makes you notice brush strokes more, being a multi-instrument musician means I’m mentally picking apart any piece of music I hear. The curse of being an artist is balancing the artistic experience with the nuances of appreciating the technical art form.
AI coding (or “Vibe coding ” - that’s how it is called) is absolute abomination .. people are generating useless pseudo-apps full of bloated unmaintainable code which is impossible to move to more pro level cause internally it’s just mess .. And what is worst, they do not learn really anything during this process, they do nit become coders.. It’s absolute dead end and soon everybod will realise it … vibec coding is one big mistake…
Inspirate rather you son to really learn to code - you can still use AI during process of learning and also during coding - but hands away from vibe coding…
I was running OpenClaw for recent week. As a huge AI ethusiast after this week i became huge skeptic. It burned 100€ on tokens and, did some funny things, broke himself multiple times to the level where i had to reinstall it completely and made me realise these things are extremely premature, basically just showcase “we did it because we can” but there is no way AI agents based on current LLMs will get massive adoption ..
it just dowsn’t work..
they are able to make “something” which looks nice at first look but as soon as you want to push it forward and make from i something “more” it often breaks apart.. And reasons why will not go away with new versions of LLMs cause thise problems priginates from fundamental principles of LLMs.
For me - this is dead end. Things like OpenClaw are like AI slop videos of coding and workflow automation. This goes nowhere.
Heh, you are a renaissance critic. I remember thirty years ago in my early twenties I would watch movies and shows from pre-digital eras and wonder how the hell they got sound effects in scifi shows etc, then after learning about how tape was used and messing with my own samples after a while I could hear 'oh that's probably glass shattering played backwards etc.' Same with BBC Radiophonic music. After a while its hard to get back the old magic feeling of just experiencing the surreal aspect of that stuff.
yah, it can be useful for some aspects of a project but it can also create a lot of problems. In games there can be prototype phases for different features that is usually throwaway spaghetti anyway. It does allow others to participate in the process more, demonstrate ideas etc. But yah it is more work for managers and leads to manage expectations, particularly for 'tell me what I want to hear' dumb ass executive types who can be too easily swayed by the surface, who also, once you start using AI assume everything was done with it.
I can understand some of that mindset. But what I totally can't understand is why you think vibe coding is "an absolute abomination" because it doesn't now, in February 2026, work as well as you want it to. Yes, if this were as good as vibe coding gets I could see it being a "dead end."
But why do you assume this is as good as it gets? Look at where AI technology was five years ago. Look at where it is now. Look at the pace of improvement. I don't see the dead end. (As one example, see Anthropic's recent project where a team of AI agents created a working C-compiler that can compile the Linux kernel. No, it's not perfect, but is it not progress? https://www.anthropic.com/engineering/building-c-compiler )
Do you seriously believe that the code generated by vibe coding will not improve over the next year? Five years? Ten years? Twenty years? Fifty or one hundred years? Why is there a demand that things be as good as you want them today, or else it's a dead end?
The claim that LLMs are themselves a dead end and "fundamental principles of LLMs" will prevent progress sounds more like a religious claim to me, I certainly don't see it. Of course, time will tell . . .
Also, I would not call the SammyJankis project vibe coding. It is a project where an agentic AI system was set in motion with some very broad and simple instructions, and a simple script. You can see all the setup files here, and see how simple it is to create your own SammyJankis. Start with the README.
https://sammyjankis.com/sammy-clone-kit.tar.gz
No. You provide help. Keep analyzing everything so we can get better answers on how things really work.
Thanks for the kind words @McD.
It makes me into one colossal party pooper though. 😂
Gotta bite my tongue every time my wife gets all excited about some YouTube video and I forget to stop myself explaining to her why its bogus or just hype. She's learned to live with me not getting all excited along with her, but if I forget to keep my mouth shut ... not good.