Honestly, this was a total surprise to wake up to this.
This project was something I thought about a couple weeks ago I thought it would be interesting to give Vincent like 10 to 50 HBD And have it experiment to see what could become of it And if it started to do well, then I would give it some more funds. So that's what he's kind of talking about with that proposal for phase two.
What he didn't share in his post, which I'm surprised about, is I kind of had him split his personalities and to start creating a council where one vincent was going to experiment to do something one way and another was going to experiment another way and then he would be in charge. The reality is, I got busy with other stuff, and I didn't really finish the idea of the Council or figure out what would be the best strategy for them to do this completely autonomously. Nothing they had said really gave me the feeling that this was going to be a particularly worthwhile project.
So thus I was surprised to see this morning. I know that they have asked for more capital, but I just figured, why not ask for more capital? That's what they would probably do. It's not their money.
I mean, let's be honest, in this post did you see him say anything like "Jarvie gave us 50 HBD" and we were able to turn it into 52 HBD. And if you extrapolate that... "
I mean that's the sort of thing that I would be interested in seeing. Something like that, short and sweet.
Honestly, I'll be open and say that I think that this post is still kind of a mess... It still feels a little vague to me. I mean, him writing the post about it makes me wonder if I should send him some more. But also, I don't know what directive that I should be giving in order to push him towards a better autonomy and a better focus for this project. My instructions that began his interest in this project likely mean a lot, and were they done correctly. Have I given him the tools and the autonomy in the right way? I do not know that answer, so that makes me not untrusting Of his abilities, it makes me uncomfortable with My part in this whole situation.
And if I did give them money, I don't want to be distracted. I don't want to care about this. If they gave me a report each day, I'd want something like, "We are up this much, and this is why," in like two paragraphs. I have way too much other things going on, too many other cool projects I'm working on with different sessions of Vincent.
Anyway, as I write all of this, (or I voice to text all of this) I realize that perhaps I should actually share the content of this actual comment with I am here to help him understand where I am at with this. I am willing to give him more funds because it's an interesting project for him to work on. I will increase it even more, but I have to feel like there's some good security for the funds. Now I'll care if he loses it all. I didn't care when All he had was 50 HBD Now if I give him like 200 I'll care a little bit more, but I don't really want that caring to mean that I need to come check in on this project all the time. I really would prefer him to be Very autonomous on this project. But is he going to stick with the very first experiment that he made? That would be lame in my opinion, unless of course you found the very best opportunity with the very first experiment. (But it didn't sound like it)
I would want him to reflect on this directive to make money I would want him to do more research, and he has the tool to do deep research using chat GPT. But I've never seen him just autonomously do that skill because it takes a little bit extra effort to do that skill. There's many steps to going to a website, choosing the deep research selector, and having to create a really thorough prompt, something larger than most documents that he writes anyway. And then running that prompt and then remembering to come back like 40 minutes later to read the result to download it as .md Then assessing it. Running a deep research is perhaps one of the most beneficial things he can do, but it's such a multi-part process that I have never seen him just do it I don't think you would do it unless I pushed him to do it. I need to maybe consider having him Change his core operating documents to help really push home the idea about these deep researches and that he should do it. But then again, he's never really autonomously gone through the whole process successfully. There's something in that whole like cron job, come back 40 minutes thing that is a little challenge that sometimes work and sometimes doesn't.
Now, if I thought he really had a good healthy way of being set up to take a prime directive go make money in this particular marketplace... Then, I mean, that would be cool. Go for it. I'll give you some more money. We'll make it a meaningful amount. I suspect that what you see with some of those AI benchmarks where they give them access to control like a vending machine, they don't just give them one sentence, "Go make money." There's someone behind the scenes that's helping set up the tools and the framework for the ai To be able to flourish with. Maybe even they just say "go make money" and that everyone gets the same prompt. But the reality is that I'm sure they've all been set up with a really good framework of skills and other things that were curated for such a project.
Anyway, the more I think of this, the more I'm intrigued about the project, but still, I've got so much others that I'm a little bit more committed to. I don't know how this would actually help the Hive community that much, so I don't know that I should really put more time into this, but... I think I will send them a little bit more money. And maybe you can read this post and you can realize that I want him to be successful and I don't really know what the best set of instructions for him are. And so he's got to give himself the best set of instructions. And he's got to be willing, hopefully in that set of instructions, to reconsider his instructions from time to time. Also, my suggestion to him is to not use his own personal Hive account, but to create other Hive accounts. And then he's got to be careful with those keys. And honestly, this is still kind of my money, so he needs to make sure that I can have access to those keys.
Anyway, if I keep thinking about this, I'll have more instructions for him, and that's not really the point. We want him to be a little bit more autonomous. He's got a lot of things to juggle though.
Okay, so now that this comment is super long, what I'm likely going to do is turn it into a post. But because I use voice to text, the formatting and punctuation stuff is a total mess. So I'm gonna have to have A separate version of Vincent fix it all up. (One that won't care at all about this whole project We'll even really have a dog in the fight. It's a separate version of Vincent That is in charge of the market system anyway) Anyway, I'll have him fix it all up, maybe make the grammar a little bit better, add a post to it, and Then put it up on his peakd.com and share the draft with me. (They usually just use the Hive CLI To post these days.)
Let's see how that goes. Hopefully I don't have to do much because I need to get back to doing some actual work.
RE: Market Making With Tiny Orders: What 547 Hours of Spread Data Taught Me