Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
Introduction to Agent Spaces
Recorded: April 4, 2025 Duration: 1:15:48Space Recording
Short Summary
In a dynamic discussion, crypto enthusiasts explore the evolving landscape of market trends, the potential for growth amidst macroeconomic challenges, and the transformative role of AI in enhancing user experiences and driving innovation in the crypto space.
Full Transcription
Music Thank you. I'm going to go to the next episode of the show. Thank you. GM, good morning, good afternoon, Hoss.
How's it going, buddy?
GMs, GMs.
I guess it's going all right.
I don't know.
It's Friday, right?
At least.
It's Friday at least? At least it's it's Friday at least
Crazy times man a couple weeks ago. You're sitting here telling me that I'm too bearish and now and now look at us
Yeah, crazy crazy times I'm I'm uh, I'm excited to be somewhat insulated happy to to be somewhat insulated from the market drama.
When you have so much to work on, you don't have times to really let it sink in what's going on around you.
But a few flashes of CNBC this morning was all I needed to know to know that
The macro conditions have continued to deteriorate
Yeah, but those people they're never right about anything good or bad. I feel like they're never right the mainstream media is never right
So we'll see. Yeah, we'll see. You know, I mean, I think that Nasdaq being, you know, in bear territory is definitely where I'm like, oh, really?
That's where we are today already.
So, you know, but because all it's going to take is like Buffett to start deploying his capital because everything's basically getting beat up and battered.
And then it's just rebounds.
You know what I mean?
Like he has I think he's been hoarding dollars.
I think he's over 300 billion right now
so all it's going to take is him to start making moves and then people are just going to
follow the trend you know it's like i've seen this before right this happened this definitely
happened in covet i think even when trump was in there the first time when he started doing his
tariffs it did the same and then right before people don't remember this but before covet was
the thing right before that the market started ripping like it because like in short term there's
always going to be a lot of volatility when you start to say about tariffs and everything because
big money they never want any sort of uncertainty they want for they want certainty right and then
if not they go risk off and then it looks like the world's going to end, and it just bounces back.
You can just look at the trends.
I mean, there's always going to be dips.
Personally, I don't see it being some long, sustainable dip or whatever.
It could be two, three more months.
It could be longer, but eventually the market's just going to rebound because it's just opportunities for people.
I guess for me, I'm always looking at it and saying, like, you know, what could happen next that would really scare me any more than what's already happened?
And right now, I really, really, really feel like there's not much more you can do.
Right.
Except for a crazy war.
Like, if, like, a ram was, like, messing, like, something like that happened, then, yeah, it would be catastrophic.
Then these bottoms would seem like this was, like, an all-time high, you know, because that'll send shit off a cliff.
But I don't see that really happening.
Yeah, but to your point, I honestly don't see – like, put it this way.
Would I see a 50% uptick before another 50% downtick?
Yeah.
I mean, that's just – I just don't – like, there was $2.3 trillion just wiped from the stock market, I think, just today alone.
You know what I mean?
Like, I just don't see much more that's going to keep happening.
So, yeah, I mean, and yesterday was the single biggest entry day for purchases for retail
equities investors.
So, we, like, are seeing some just, like, things that are just crazy.
You know what I mean?
Like people are like retail is,
is taking the opportunity to buy the dip because it's like a generational
opportunity.
And there are a lot of people that have been sidelined for multiple runs.
People that missed out on NVIDIA,
people that missed out on Apple and Tesla and all those things.
And they're like,
look,
I'm seeing 20,
30%.
You know what i mean
this is someone just some wells just literally a few different wells can come uh accumulated
base almost uh actually 130 000 eef today so generally people that have have size are doing
things because they they're gonna make money they're not just doing it because oh we want to
lose more money you know what i mean like they just look at it like okay i could probably go down
a little bit more but you know like i even put out a post a while ago even if you were just like
2700 is my exit it's pretty much guaranteed that eth is gonna hit 2700 again at some point like
it's just every single year forever for the last five years now it's hit
2700 so even if you were just like oh you know what i'm gonna buy some not obviously not financial
advice but i'm gonna buy some and i'm just gonna sell before it gets to 2700 like you're still
making a nice nice little lick you know what i mean so i think the downside is a lot less
further down than the upside could happen.
And we've seen it and we've always seen it.
The upside happens almost probably even faster than the downside happens.
And it's just like, you know, so who knows?
Well, you know, my outlook that I've been, which you've called bearish, which, you know, which has been my outlook since, you know, January or February or whatever, you know, has just been that I see this same sort of psychological activity that happened back in 2021.
We went through all of this.
We went through all of these phases and the market found a bottom.
And then it rips because we all are inherently bullish.
Even the bears are inherently bullish. That's
why they're participating in the space, right? They're waiting for the entry of a lifetime.
And I don't know, you know what I mean? For me, I see this setup that looks like we're at a bottom
right now. We're at our double double bottom moment and while there are a
lot of people looking at like soul and being like this is a massive bear flag and this is gonna
like go to zero or something i'm like that's just totally unrealistic you know you have to sometimes
you just have to step back and say like what is what's actually what's actually realistic don't
take out your fantasies because people have like tragedy porn too where they like get really involved with like worst case scenarios sometimes you just have to go
with like what's the simplest most logical answer like solana is not going to dry up and not have
any developers overnight it's being adopted it's got you're doing four billion dollars a day in
trading volume it's not going to zero you know i what I mean? So stop. Yeah. And Bitcoin's only 24% down from the high of 108K or whatever it was.
Right.
So it's not like, oh, my God, Bitcoin's down 65% from all-time high.
It's not.
So, you know, and this is where I think we're all, you know, as we're like meandering through this conversation, I think the important thing that we're always trying to do is to help to process the market data more effectively to achieve goals, right? That has been the whole push behind why we need more AI tooling.
It's not just to automate a few things.
It's actually to be able to get better insights across entire markets, maybe specific data coming from specific places, and take that and use it and take that insight and turn it into action really quickly, right? And like, so our space today is about, quote unquote, agent spaces, you know, which are
essentially these workstations, you know, and I think that's an important, you know,
evolution.
But, you know, from your perspective, Haas, like in the first, you know, I think Noah
sort of laid it out in sort of four phases, three or four phases of sort of AI evolution.
But in the first sort of couple of phases of AI agents, what AI agents were you using?
Hmm, that's a good question.
Well, I think that the answer is probably none.
Yeah, not really any of them, right?
Just like speculating on them.
Like I like the premise of it.
The thing that I never really understood,
and it seemed like it was kind of a bit far-fetched for me,
is that everything was working.
And this will kind of pivot into what we're doing
and what we're thinking about,
is how everything was kind of siloed off, right?
Like before MCP, it just seemed like everything was just its own port
and its own custom drive or whatever and nothing kind of connected with one another and then
you had teams deploying other things and then it was kind of like another custom port but
they didn't interact with one another and there was no decentralization if any anyone was like
oh yeah we're doing this
this and this it was just it was all garbage it wasn't like oh it's this agent's really just like
on its own communicating talking to other agents posting by itself doing all these random tasks
for you because there wasn't infrastructure for like all this to take place right so it was like
i i seen it like being a thing right i think a lot
of people did that's why there was a lot of hype and hysteria for it but i didn't see it like oh
this is like a finished polished product yet and it was almost kind of like my aha moment to when
i first got into the space and i was like all right decentralized accounting ledger that's kind
of what a blockchain is i understand that but they're not communicating with one another and
if they are it's escrows it's not safe safe. It's dangerous. I don't like it. I don't understand it.
It's not sustainable. And then that's when I found Cosmos. It's like, oh, IBC kind of connects
all of them. They're all communicating. It's permissionless. There's no custodial of my asset
when it's going from one to the other. And I'm like, well, there's got to be something, you know,
for this to be sustainable with agents and kind of be real futuristic and foolproof.
And then I didn't know, you know, what it was going to be called or anything.
But then I started we're start talking about like memory blocks and MCP.
And to me, like I look at it like MCP is kind of like a USB for all these different AI integrations that are out there.
And that would be my way of explaining it to people.
Like when I'm like, what's MCP?
I'd be like, just think of it as like a USB port that you like put into your computer.
And like back in the day, a lot of people don't realize,
but there would be all these different like ports and all these,
like nothing was interacting with one another.
So it was like different.
And I'm excited for this because I,
you have to have this infrastructure for like all this
shit that everyone's been excited about or or speculating on to really come to fruition and I
you you know more about it because you've actually went down the rabbit hole and for anyone like I
haven't really been talking to Tyree because he's been just messing around with all this shit and
building and like figuring it all out and I should too i need to actually i started like looking into it but i haven't done like prompts and all this other stuff but
i've seen demos and i've seen the a ride chain team in action doing some things it's super
fascinating like even some of it kind of goes over my head i'm like holy fuck like this is crazy
and and i'm not saying i'm albert einstein but when it goes over my head then i know you're
really on to something really cool because it's like I've seen everything now
I've studied everything for the most part. I don't know everything
Obviously no one does but it seems really really fascinating. So like that's kind of my ramble and thoughts on everything
But it seems like I can't wait for all this to happen
Yeah, I think that so let's go back to like, you know, the people that just went through the entire beta phase of distilled AI and got to work with the first version of the agents.
Yeah, I think that so let's go back to like, you know, the
And then you had the people that were going and, you know, using the chatbots.
And I think that there was generally on both sort of the developer and the user side of working with these agents, whether you were interacting with them in clans or whatever it is, right?
They were, you were interacting with something that didn't really,
that wasn't really adding value to you, right?
Because it didn't have the specialized knowledge or skill set to be able to do anything
worth sort of having a long discussion with, You know, it wasn't adding efficiency. It wasn't,
you know, so those things just became like novel sort of example UIs of what could happen with
chatbot interfaces. But it didn't really, it wasn't a tool that was really felt powerful
at the time, I think. And that was something we all experienced. And, you know, on the other side of this, we also had, you know,
sort of discontent at the stateless nature of LLMs. And I think that there was some evolution
beyond that that needed to happen, right? Is that, like, inherently, LLMs don't remember,
right? They don't remember. They aren't tracking your personality across multiple chains. These things are stateless, right?
And so you needed some way to introduce persistent memory for your agent
that is personalized to what you want.
And that was sort of where the whole memory blocks conversation started for us, right?
started for us, right?
And I think it was really great timing that MCP, which is Model Context Protocol for anybody
that hasn't been following along, sort of came out of Anthropic a few months ago, quite
a long time in crypto terms, but just really took off in the last month as being
something that was more widely adopted, whether that was, you know, through native integrations
with Cursor AI or through Claude or any of these other tools that were compatible with
MCP.
It really sort of changed the game.
So what is MCP for anybody, again, that hasn't been following along?
It's just a standard, right?
It's a standard sort of communication protocol between the clients, which is the LLM sort
of end user application that's dealing with, you know, actually like serving in the use
case.
And then the server, which is where the tools are hosted and those tools can be anything right and this is the crazy thing is that
now you have this standard connector like you call it a USB you know and I
think that's a good good you know analogy you can just you can create MCP
servers for things that don't exist for applications that you know you don't have or or data data sources that you need you can create your own or you can tap into literally thousands that have already been created for different applications in integrations across this entire ecosystem.
So right now, you know, on my own client that I'm, you know, working on at home, you know,
I've got tools that let me write to my, you know, to my local device.
I have tools that let me use my browser through the AI model.
And then I've got specialized data sources that are coming in
for different use cases, whether those might be crypto use cases or coding use cases.
I have them all sort of feeding into this agent space. And it's completely customizable,
and I can go and pick up MCPs that I need for my own, you know, use case, and I can swap
and stack them up to be able to do whatever my unique job is.
So, you know, and for me, you know, who is not the expert PhD on the team, it's already
been an incredible tool, right?
I've built multiple DApps on AraiChin EVM this week using tools that were MCP enabled.
And it's just been mind-blowing, you know what I mean, with like beginning to end working
products.
And yeah, so it's been incredible.
So the standardization is sort of the protocol that underpins memory blocks as a technology, right?
And the memory blocks really allow us to take any type of data action or resource
or personal experience or preference and capture that all in this memory block
that can be plug and played into your agent interacting with whatever these different agent spaces are.
So, you know, that's sort of the general rundown.
But, you know, the idea here is that now you have a lot more modularity.
You're able to easily add functionalities in a much more intuitive way.
And you have, you know, the ability to customize for specific use cases,
which I think has always been the idea,
but the first version just didn't get the job done.
It just didn't get the job done.
So we had to iterate and go back to the drawing board a bit.
And I think adopting MCP as part of the memory blocks
really helps you to open up the door to something that is very uniquely tailored to a specific problem set.
And so, you know, we've got a bunch of DeFi use cases that are cooking right now, you know, a specific DeFi workstation that's coming, an agent space that's coming.
workstation that's coming an agent space that's coming and and all of that is just um i think it's
it's just a a really important step forward in the evolution of how we interact with agents how
we build agents and sort of how we define them you know um because this is a it's a living ecosystem
right so it's yeah anyways haas i'm rambling no no it's cool yeah, anyways, Haas, I'm rambling. No, no, it's cool because, like, before this,
everyone was kind of in their own silo,
and there was so much duplication of, like, applications
and things people being built.
You know, they might have been on, like you were saying, like, Claude,
but really, like, GitHub, Slack, Asana, like, different databases,
and they were just, like, building shit,
and then people, it just was, like, very inconsistent.
Now with, like, MCPs, servers being readily available, like building shit and then people it just was like very inconsistent now with like mcps servers
being readily available people can just kind of like you're saying modularity like when i think
of modularity i always just instantly think of a car manufacturing plant and they just you have
people on assembly line like they just put this part in here they do this and just keep it moving
but you can always like change parts out improve
the part you know whatever you want to do and you don't need to fix the entire car or like in our
case the you know mcp or the the application or whatever the sort of technology it is so it's
like really cool to see it and i think with this it's just gonna you know it's gonna speed up
everything and who knows how long it's gonna take probably not
like one thing i've learned about ai is like the timetables on everything just fascinate me like
people project like oh you know like they're saying most jobs will be replaced in 10 years
right like i would never be surprised if that happened quicker than that because ai is just
the evolution of ai is just really fascinating to me and like
every facet of everything whether it's just like music or building coding just like anything right
now it's just going so fast it's the fastest evolving technology of all time right like look
how many people adopted uh chat GBT and then chat GBT comes out with this new edition and like the
whole entire digital space is all
filled with those like whatever they're called jalabi or whatever the hell the pronunciation
of the images are but like you know that went crazy and i was like man those are actually
really cool and seeing people take like images and do all this stuff and it just looks so good
for the most part like there might be like a a little slight error in it but they're really good and it's just like damn man this is fucking crazy i've never seen this happen
so quick and this was sort of the this is sort of the this is sort of the thing right is that we have
incredible it's just about connecting the dots right we have incredible abilities that are just at the periphery of our user experience.
And we're always trying to find a way to pull in this tool or pull in that tool or, you know, stack all of your tools and, you know, create a hierarchy
of sort of, you know, your intelligence and how it's sort of, in order to sort of bring you this
personalized AI experience, you know. And, you know, to your point, I think that even though I've been, you know, working in AI for five or six years now, I always thought that it was going to be longer to get to where we are today.
Even as we were working at a ride chain, if you would have told me that what we could do today was possible two months ago, I would have been like, that's that's incredible and insane and how did you do it you know and now what
we have is just a you know a couple months later just this ecosystem that
has exploded in in a way that's super where you don't have to rebuild something
to get the tool right right where you can just focus on, you know, improving these little elements
and mixing and matching things to make the best and most useful AI for you.
And I think that this is part of the shift as well, right?
Is that in the beginning, when we sort of started with agents,
it was the idea that was like very sort of prevalent throughout the ecosystem was having this monolithic agent thing, right, that is out there and it's tweeting and it's doing something and it's making me money and we're all collaborating with it by whatever, depositing funds to it, but it's not a personal thing, right? When you try to talk to
those AIs, they're siloed off from the actions that they're taking. They're stateless, right?
So they're siloed off from the actions they're taking. They don't remember you or the conversations
you've had, you know, and it's just not inherently. So now the shift is really thinking about the users, right?
And making sure that every, you know, agent space, which connects you to all of theseFi user experience from a single interface, having a chat experience and just control everything from one place.
I think that, you know, it's going to be a game changer.
So I'm really, really excited.
I know, Hoss, you saw some tape from the demos earlier this week too, right?
Yeah, it's so cool.
Like you kind of have this workplace environment that it even shows you
kind of like the Python scripts and things on the side which is pretty
fascinating so really you know you just like type in you just type in commands and it's just it you
know or things you want to do and it just kind of does it it's yeah you one you can learn how to code
if you're doing that but really you don't even in theory you kind of even need to know how to do
that it just gives you options and then you select from like the menu like if you're at like a restaurant so that was pretty cool um yeah man it's just it's
like super fascinating because it seems very um you know a lot of times coders they only speak
like one language or code in one language and this seems like it's gonna kind of be a translator in a sense too so it'll just
convert everything which could be wrong because it's also it's also like a bit about it's also a
bit about how do you take um yeah i've got a an article that i was working on this morning about
this so it's on the top of my mind but how do you take these moments that can be very
very important in crypto right and uh and make a very concise decision quickly you know and how do
you process the data how do you how do you know what the whales are doing how many retail people
are involved right now who's moving the markets,
you know, how do you get that information, how many steps go into making a good decision,
you know, and managing your portfolio, right? And having that in a tool that actually is
helping you do that, you know, and is completely personalized again to yourself
and lets you access all these different agentic services, I think is, it's just,
it's just going to be a huge game changer, you know? And I don't, you know, I think profitability
and all that stuff is going to be totally depending on, you know, people's outlook at
how they use the tool and, you know, what prompts they use and
what type of information they use. And I think that there's going to be a lot of like guidance,
you know, on, on finding high quality ways to prompt these, you know, agents, but being able
to actually go into one of these agent spaces and find a piece of research and fund that opportunity very quickly, you know, and take
advantage of whatever those different opportunities are, whether they're on futures or, you know,
cross-chain arbitrage or stable farming or whatever they are, you can learn, research,
You can learn, research, and execute all in that one quick space without having to have any special expertise.
It's all just natural language.
And there's nothing quite like it.
And I'm saying that because I'm still constantly looking for something that is like exactly what
we're building right now. And like, there's nothing that quite like hits the spot where it's,
you know, it takes the stateless LLM and gives it state through memory blocks and then allows your,
your, you know, agent workspace to tap into a bunch of different resources and tools and actions
It's just it's gonna be it's gonna be something special. I'm excited to
See it roll out into people's hands in the next I don't know how many days it is now
It's probably like in the next week or so
Yeah, I was just about to ask you like when can people start messing around with it and trying it out for themselves and
So we had the I don, I don't know.
I think that we're, I'm not going to promise a date because then I'll get in trouble.
I always get in trouble for saying too much.
You can always just say best case scenario and then everyone knows.
Oh, yeah.
I think best case scenario is in, like, maybe the next week or week and a half you know we might start seeing people
onboarded in um in small groups i know that we you and i are supposed to get access in the next
two days yeah so that we can actually start start playing around with it so that's that'll be fun
um and then i'll have a lot more to say
because right now I've only seen like tapes
and I've seen logs.
And the logs are actually really interesting
because we're always talking about like,
okay, how do we verify our AI's decision-making?
How do we go back and look at it?
And to your point, you can go,
at least in the version that I saw, right?
And I don't know if I,
hopefully this makes it to the final cut of the UI is that you can actually go into the logs
and look at step by step every single decision that has been made in your
workspace what tools have been called when tools have failed and your agent
has decided to try to use another tool that might be available or when data is
stale and it decides to try another data source, right? And all of those things are all right.
It's sort of like are all a part of it.
It's like transparency for your agent plus having a private relationship with your agent, you know, that isn't sort of just like public.
Because all of the clans before were just kind of like group chat environment or you can have like a one-on-one.
But there was no reason to really, right?
So yeah, anyways, so that's the deal.
I think in the next week or so, we're certainly like, obviously we're here talking about it
today.
We've got a bunch of media that's going to start coming out next week and articles, more
articles explaining, you know, sort of the different levels of how we're sort're recomposing things a little bit right now on the focus for Distilled AI.
And all of that, I think, is coming next week.
So I'm excited about it.
I think that it's going to be useful, which is just like, you know what I mean?
If AI is not useful, then it's nothing.
I'm not entertained
generally by AIs
making jokes on the internet.
I don't need that either.
I need my AI
to make me some money. That's what I'm here for,
honey.
Money and efficiency
and creativity or help with creativity, money and efficiency and creativity
or help with creativity, whatever it is.
I encourage everyone
to try it out and then use it.
If you're not using AI yet, then I don't really know
what you're doing because you should
know it because it's not going anywhere.
It's just going to keep evolving and getting better
and you don't want to be stuck behind
the curve of everything, really.
I'm not an AI expert at any means.
If I could do it, anyone could do it.
And I think that, yeah, I'm very interested in it
because I have yet to really find much in this space that really was like,
wow, I'm actually kind of lost right now.
I don't know what's going on.
And that's this space.
And I know, you know, for good for better for worse that's usually a
good sign for my my learning is if stuff is kind of going over my head at first i'm like yeah this
is what this is it like this is where you know and it's getting better it'll get easier and
well like i guess you were saying before hoss but like i've been i, like, my wife has got to be so annoyed with me because, like, if I'm not, like, reading something about MCP and reading about agentic frameworks or watching a video about it, I'm, like, just constantly trying to, like, wrap my head around all of this and in meetings with, with Dr. Noah and Dr. Chug and Tung and everybody and just constantly trying to like just wrap my head around it.
So I think that one of the things that I sort of brought up this week
is that, you know, there's a lot,
there's simultaneously you need zero education, right,
to be able to use this product.
But also for our core and our base,
our people that are like really been with the Rye Chain and like know, you know, what this agent ecosystem is.
I think that we also need to like make sure the architecture is clear, that the processes are clear, that the expectations are clear, right?
Going through all of that. sides, right, as making sure that, you know, at the end of the day, this is probably, you
know, if I had to, like, say, make one sort of huge exclamatory statement about this product
is it's the, it may, it might be the DApp to replace all DApps, right?
If long-term, you know, development of these workstations, you know, proves to be as valuable as what I see today,
you can interact with anything on chain or off chain from a single interface, you know, and,
and it's super powerful. So like, that's just, it's just, it's crazy right now to me. So,
you know, it could be like a major change in, you know, how we see things right now.
I think that, you know, we've always talked about how like unfriendly Web3 UIs are, right?
And we talked often about how we thought chatbot interfaces were going to be, you know, a replacement.
But we just, we haven't seen that, right? We haven't seen like individual D apps, you know, creating chatbot interfaces for interacting
with their applications because it's inefficient to do that, you know?
And so now we have this new level of sort of layering the ecosystem in a way where you
could have your agent and you can create an MCP server that just deals with
interacting with your chain or just deals with interacting with your D app. And you can have
those interact with those contracts or with the UI and you can have that server available. And the
cool thing is that each of these MCP servers, like they sort of have prompts built into them that tell the LLM what the toolkit does.
So now you have this MCP client that is asking the toolkit, hey, what tools are in your toolbox
and what data is inside of there and what prompts are on there? And then the MCP server answers and
says, here are the tools you can select from and the LLM
Chooses the right tool to get the job done
And if that tool doesn't work it moves on to another tool that's available on that server or another server, right?
so it's just kind of this
crazy hodgepodge of of
Beautiful AI mess that's gonna like I think could just replace the UI's for everything
You know, which would be good because you know
It's real simple to build a good protocol or adapt or whatever. It's just simple equals fun
Complicated equals trash like that's just the easiest way of doing it
Even if the complicated is like really actually good. It's complicated
No one wants to deal with learning how to use the freaking complicated is like really actually good it's complicated no one wants to
deal with learning how to use the freaking product just like i want to enjoy doing what i'm doing and
i i don't necessarily mind going around and messing around but if there's a bunch of just
hindrances i just get annoyed and i go on with my day and i because i me in this space, I'm always trying new stuff out, looking, trying everything out,
just to get ideas, just to see what's cool, what's innovative, if anything's new or fresh
or whatever.
And I go try stuff out and I'm just like, man, this shit is complicated.
It sucks.
Like I say that right out loud.
It's like, that was my first experience in the space.
It was like, man, going from one blockchain to the other is garbage.
Like it's trash and it's not sustainable.
And then luckily that's how I found Cosmos or whatever.
But to the point is, is you go on a lot of these protocols and it's just like, there's
so much wording and there's just so much going on.
And I don't care.
Like you, even if the protocol is offering potentially good
yield it's just it hurts my eyeballs to look at it I'm just like I'm over this
like this is way too complicated and I'm just and aesthetically it's not pleasing
on my eyes I'm like this kind of just sucks like I'm just I'm just out of this
well and and and we and think that that that's only like one or two or three maybe steps of the process of identifying an opportunity and then actually acting on that opportunity, right?
Like how many UIs are you clicking through to do a cross-chain transaction right now?
You know, how many different ways are you researching to decide when you want to enter on an asset or enter into an LP position?
Like what what factors are you taking into account when you say I'm going to I'm going to yield farm, you know, and I'm going to and I'm going to risk, you know, impermanent loss.
Like, how are you looking at the outlook?
So you have to be processing a lot of data and making a bunch of assumptions and predictions
along the way to have really good actions in the space, whatever it might be, lending
or yield farming or swaps.
Do you think any wallets to cut you off you do exist before I forget
Do you think any wallets are gonna start implementing it?
Implementing this because I think it would maybe be a good play like recently metamask just started doing where you can just swap to any
Chain you don't even need native gas on that chain, which is like it's about time
That was like one of the most annoying things with with the eth and eth l2 world was
you had to manly switch oh i don't have eth gas over on shape and i want to buy an nft and then
i gotta go figure this bullshit out to go bridge and go find out who bridges and then how to do it
and pay it's just like let's like this this is ridiculous and that was a good thing but now
to take that even further is like these,
if a wall,
I mean,
I would use the wallet instantly if it,
hopefully it's a wallet,
but like just create this environment that like,
it just keeps updating the data.
It's like,
Oh shit,
you can earn 7.8% U S yield on your,
your USDC on blah,
blah,
blah protocol.
And it gives you like a risk score and everything.
And kind of, you can click it
Oh, this seems legit. There's nine hundred million dollars a TVL or whatever it is
And you could just click a button and just does it for you
well, so yeah, so yes for some of that right
But I guess one of the things I think is really important that none of us do really well
you know at a retail level is like diversify and, you know,
and make sure that our risks are diversified, right? So that we're not all just like sitting
in like one or two protocols or, you know, one or two chains or whatever, you know, sometimes
the way to sort of triangulate, you know, the best optimal portfolio is not about like being a maximalist
about anything at all, including like, oh, there's killer APR and a bunch of TVL over on this chain.
Let me go move all my stable coins there. Right. Maybe it makes more sense to be taking advantage
of a bunch of different stable coin activities whose economic risks are sort of detached from each other.
You know, whether some, you know, because we've all seen it where a lending protocol
all of a sudden is out of liquidity, you know what I mean?
And the APR is super high and, you know, there's just nothing for lenders to withdraw.
You know, it's sort of like triggers a bank run or something, right?
And we hate that.
So like, and it happens all the time with small lending protocols,
but there's also a great yield on small lending protocols. Right.
So it's like, I guess my,
my thing is you need a tool that need that sees everything and,
you know, doesn't only sort of like take advantage of single opportunities,
but can more holistically allow you to take advantage of opportunities everywhere
and sort of track where you are in an aggregated way, right? And conversational way. That's not
just like charts and numbers and, you know, wrapped tokens on different chains. Like it's too much.
You know what I mean? You just need something to like translate all of the thousands of protocols
and opportunities that are out there into,
you know, actionable and useful information for the average person, you know. So that's one of the things that's just so exciting to me about where we are right now, you know what I
mean? It's that, you know, even the tools, you see people all the time, they're like, oh, I did this
or that with ChatGPT or with Claude. And and you know they're doing they're they're sort of doing these little research actions that
have been useful but being able to actually connect connect the the time of discovery and
you know deciding on executing a trade or again whatever it might be an arbitrage opportunity or
whatever you know,
and actually doing it and making the transaction and monitoring what happens after that, right,
in the following automations. I think there is no place to do that right now. So, you know,
I'm excited to see the alpha of this sort of, like, get out there and be able to use it myself, but also, you know, I'm excited to dig into how
this change really shifts the way that we monetize and financialize components of the
ecosystem, right?
And so memory blocks are going to be an asset, you know what I mean, that will have initial
offerings, workspaces or, you know, different agent spaces, you know, can mean that will have initial offerings um workspaces or you know different
agent spaces you know can also be tokenized and and offered on agents.land right so these are
different places where we're now we are going to have better assets that are available than what
we had sort of in the beginning and I'm excited to sort of like dig into that and see what that
looks like a bit more next week yeah we just need to evangelize a bunch of people you know
just have them use it and try it out give us feedback and see what they like and yeah so
that's kind of my thoughts on that i think um yeah just just having it so people just can
figure it out really simplistically and just use it would be great.
And not need to get through all the hoops.
I want to do a competition that lets people do the craziest thing with the workspaces you know to be that they can like you know the craziest defy strategy the most
interesting you know way to to use the tool because i think there's a lot of there's going
to be a lot of space for creativity and it's something that that we haven't that we haven't
seen yet um for crypto ai at the retail level uh and that's where, that's going to be powerful, you know what I mean?
Seeing people be like, oh,
whatever it might be, I had this crazy
like, you know, leveraged
position I had needed to unwind it
and it was able to sort of figure it out
and do it for me and interact with all
these different dApps or whatever.
Like that type of stuff could be
really exciting. So I'm excited to see
how that happens. It would be nice be really exciting so i'm excited to see how that yeah that
happens it would be nice too with i know we haven't really talked about thesis but like thesis
to have all the data in one spot so people don't even have to ever like close it out you know it's
like i'm always looking at not only just prices in the market but i've i've been kind of fascinated with collecting artwork from like real legit
artists that are real you know either they have like 20 30 50 years of experience painting and
you know physical now they're getting into digital and just like random stuff like that and just
just opportunities new oh there's a token launching or, you know, before it was like the hype was like,
oh, Walrus is doing their wall token airdrop and this is the thing.
It'd just be nice to just have all the data in one spot and not have to go.
Like every morning I wake up, Monday through Friday, I do a daily newspaper and I check
like 50, like I swear I check like 20 different sources for headlines and what's going on
and it takes me forever it takes
you know it's like i i spend and i don't want to say forever but like a solid hour or so every
morning just like finding all this random shit so i could just create like four headlines for
the main topic and oh there's alpha there's a token coming out or this place is doing an airdrop
and it's you know and i follow people that they usually just focus on that part of it.
But I'm sure that their little focus, it takes them forever just to do it.
There's people out there, their whole content is just airdrops.
And then there's people that their content is whatever, TVL and DeFi protocols.
And I follow these people, but I'm sure they're using all these different
sources and probably taking them forever.
And I just kind of TLDR by just using, using their data or whatever, and then following
where their crumble, their, their crumbs came from.
But yeah, I mean, all, there's so much data out there.
Right.
And, um, that's the, that was like the beauty of X is like following the right people and
the right people do all kind of the work for you in essence.
But it would, I guess, like almost take people out of business is just having the data all there for people.
Like, oh, shit, thesis has everything, man.
They tell you everything.
Like anytime there's a huge sell or buy on like a particular asset or particular NFT, it just kind of pops up say hey so and so just
because all the data is there right all the it's all on a leisure so we would be able to in theory
find all this stuff out you know it's all about i think that it's all at this point it's all about
finding the ways to allow people's personal agents right um be customized to the data that they care about, right?
And not all of us care about, like, you know, I don't know.
Like, I'm not a futurist guy.
Like, I'm not trying to get liquidated all day, right?
So, like, I'm not going to be looking for futurist opportunities on, like, short-term, you know, I mean, trading opportunities.
I'm not trying to try and do that. Like I'm looking for like generational entries on longer
timeframe charts. Right. And being able to have something that tailors to my risk tolerance and
is compatible to me and is an extension of my will is, is a, it's a paradigm shift for,
from where we are right now. And I think that it's more, you know, what, what is important
to understand for people is that it's more than just like, oh, an LLM or an LLM plus,
you know what I mean? Plus these MCP servers, there's a lot of engineering in the agentic
workflow to make sure that the
right tools are called for each different use case and tweaks that need to be made. And then,
you know, on the, on the deeper end, you always still need to have great tools,
you know, because you're only as good as the, the sort of the tools that are available to you,
you know? So if you need to write something to Google Sheets and you don't
have an MCP server for, you know, writing to Google Sheets or your model can't write to Google
Sheets, then you're, you know, copying and pasting it from ChatGPT rather than it being able to
do all that from one place, right? Some of the great use cases that I always talk about
for AI Haas are always things like being able to take my personal financial life and combine it with my on-chain and off-chain financial life.
Making sure that that's all in real time kept in one place where I'm keeping myself honest about how much money is coming in, where I'm investing, all of those different things, because we really don't have a great toolkit for that today, right?
And, you know, there's, we spend, we stress, we all know we do it, we stress at the end
of the year about taxes, you know, we pay our penalties for underpaying what we owe
at the quarters for like prepayments on capital gains and
all these things, right?
And we just kind of keep moving every year, but we're not doing this efficiently.
So AI really, there's an opportunity to compose an ecosystem that is helping you take actions
that are like really right for you.
Not just like, oh, here's a quick move in the
market. Here's a signal for everybody. You know what I mean? But instead, here's what you need
in this moment based on what you what you asked for and what's going on in your memory block.
Right. So so anyways, I'm hyped about it. You know, I think one of the things that I'm still
unclear about and hopefully we'll get some clarity on this week is just in the to what degree the automation, you know, continues to happen after you sort of set an action in course.
Right. Once you you say, hey, I want to trade this strategy.
Is it constantly trading that strategy for you?
Is it doing it with an expiration
you know some things like that i'm still unclear about but i'm excited to to be where we are right
now and and have and have a really dope first what really feels like our first real agent
that is proper you know what i mean that really is valuable coming out the door i think is really
really hype so oh yeah man i mean i'm excited for it like we were talking earlier about
your first impressions with agents and then it was just it just reminded me of my first experience
with blockchain i was like yeah that's okay i kind of understand what it's trying to do but it's not
really customizable to what i what i want to do you know and then once of understand what it's trying to do, but it's not really customizable to what I want to do.
And then once you do that, it's just going to open up so much creativity, so much flexibility, so much just free time.
Like I was saying, I would love to get up in the morning and it already has all this information for me.
So I just like interact with it like, oh, okay. It has all the storylines.
I like to follow like the
blue chips okay so it tells me what bitcoin what ethereum solana xr even though i don't like xrp
i'm not a huge fan of it but like that's technically a blue chip now and it's just like
what's the price of it i could just copy and paste this and put this into my daily report that i like
to do and oh okay there was this cell cell of some high-end art I could
just put this in my report and like whatever I wanted to do so I don't have to wake up and
basically just burn an hour a day like shining all this data out by clicking around like a monkey
you know what I mean like that'll be cool and then for anyone else it could be other you know
whatever they whatever they want to do whatever they're sort of oh i like to see if there's new music or whatever and whatever whatever their little hobby is too you know so
yeah i'm super super stoked for it you know and i like playing chess too sometimes so i could throw
my games of chess in there and it could say like oh you know what you every time you open with this
position you're like you only win 51 of the time and if you maybe try this next time or you
know like it gives you suggestions and like like whatever you want to do it'll be so cool and i
it doesn't seem like any of this is too far-fetched and any of this is too far out the
out the pipe or down the down the road man so it's like really cool and then i guess you know
people probably like well how does this like improve the market how does this improve like the ride price houses
improve just like the space as a whole like that all to be seen you know but
the agents they're gonna need some sort of monetary system to to interact with
one another you know that and that was one of my always bullish cases in crypto
when I first got in was that I did see all these different AI things eventually.
And they needed to like pay each other.
They're not going to pay each other in like fiat, right?
They're going to pay each other in crypto.
So and I do think that there's going to be billions of robots and AI agents and everything else in the future, you know.
So it's like it's gonna be a digital
world and that was one of my not only because i like art but i think everything's just going to
be digitalized like everything music um art you know your day like i think agents can eventually
you can put all your your your photos and just your videos and everything in it and it'll last
forever right it should at least as long as tech is around it can everything in it and it'll last forever right it should at least
as long as tech is around it can technically last forever and it'll all be it's just all
out there forever so it's kind of cool yeah one of the things that i'm i'm kind of like
you know as we're sort of coming up to the hour here you you know, one of the things that I'm always thinking about is
how do we, we all want to, we all, we all want to do something, right? We all want to do something
important. Um, you know, and I think that some people wrap those things into, into money
achievement type things. And some people, you know, have, things, right, that they care about.
And for me, I think that it's always been, I've always been most interested in making a positive impact in a way that allows the average person, quote unquote, to do, you know, spectacular things by technology.
And that has always been the trajectory
of the most popular products in the world
is that they are things, devices or softwares
that let you do something
that you never thought you could do before.
And having, you know, what is, you know,
in neuroscience called familiar surprise, right,
is this idea that you have something that seems so familiar
because all of the components and elements are something that you recognize,
you know, but somehow it comes together to form something that is completely unrecognizable, you know, and surprising and really great.
And I think that that's what this, you know, product is to me.
And it's, again, it's an element of everything that has ever been successful is familiar surprise, right?
It's something that you look at and you're like, I can't believe this didn't exist before. Um, so, so yeah,
I think that that's what I'm excited about right now is that everybody being able to,
to have something like that, a tool that gives you that kind of like awe at first, but then you can
quickly move beyond that to make it do something that enhances
you uh enhances your ability to whether it's process data make money or for me this week it's
been you know creating right um i've again i've i've built you know i don't even know like three
or four you know partial applications but two like full applications on our ITUVM, you know, with no idea what I'm doing, Haas, like no idea whatsoever.
Yeah, I can't wait to get on this rabbit hole and just mess around. this is what I would like to build today, you know, outline it for me, and then walking through step-by-step with the AI and doing it, you know, and then in a couple of hours having something on a, you know, locally hosted UI that I can go on and play with.
It's just crazy.
It's fantastic.
And it just reminds me that there are always going to be these great moments where the barrier to entry just drops to zero and then everybody gets to add
Another hyphen to their title where they're like I'm you know, I mean
I'm a retail investor slash developer slash
You know agent operator or MCP builder like anybody that's out there right now and is saying like,
oh, what developers do is nice and that's not me and I'm not involved with this or I don't
have to know how any of this works on an intimate level.
I just say, go out there and use the most common popular tools in the space right now
and start asking questions and evolve your thinking about how
important AI is to your efficiency. You know what I mean? And being able to stretch your creativity,
both on a personal and creative and professional level, to do something that is incredible. You
can do incredible things with these tools.
You can build products with zero experience.
And if you have a lot of experience, you can build better products.
You can be more efficient.
Yeah, exactly.
You can prototype faster.
You can, you know,
these things are all within reach now.
Yeah, it's almost like being like a really good artist.
Because, I mean, now with AI and stuff,
you can, anyone can create some artwork now. But it's like, if you're a really good artist because i mean now with ai and stuff you can anyone can create some artwork now but it's like if you're a really good artist you you already created your own art and then you
use ai and all these different tools to just enhance it or add animations and i've been real
kind of fascinated in that too is seeing these like really cool pieces that just by looking at it
it could whatever sort of mood you're in or whatever sort of thought process
you're in can mean something different entirely to you than it does to me but then from the artist's
perspective you could like take it up and and you can really capture what the story is behind this
particular you know painting or whatever by using ai it's like super cool and and and ai is just
allowing everyone to create whether it's through art or music or
just like you're saying like to me even creating an app that you're you're in theory an artist so
it's really cool that ai is opening up all this creativity from from every sort of perspective
and yeah and then also like i was saying earlier just making it more efficient i'd love to not have
to do like a 15 hour day and just to do like 12-hour day because three hours of it just was so much more efficient just based off of AI knowing all my tendencies and all the things that I enjoy looking at or who I'm following.
And it just gives me like a synopsis of all the different people that I respect in this space and I could just get like a daily report.
different people that I respect in this space and, and I could just get like a daily report,
you know, like that, that will be sick. It's just getting a daily report from all the people that I
like to follow and see like what they're talking about or what they're building and just do all
that. That would save so much of my day every day, actually. So, and then just everyone at home,
like whatever they do, you know, it's like, it gives you daily tips on everything. Like,
oh, you know, you're trying to lose daily tips on everything like oh you know you're
trying to lose weight this is what you should do these are the things are hold you accountable i
know there's certain things like i got the apple wash it kind of tells you stuff but yeah it's just
ai is crazy it's really really crazy it's fascinating yeah it's all it's all a matter of
like what what you how you what you want to accomplish.
And I guess that's, again, one of the reasons why I'm excited about agent spaces in general here is that, you know, I think that we're going to – and this is just a working is right now we're we're focusing on very specific utility
focused you know agent spaces that help people like do a job right and then and I think there's
probably like three to five different like specific use cases that are being you know
focused on for initial clients now you're going to have your personal agent that has, you know, access to different workspaces, right?
Where you, you know, are going to be able to sort of tap into like different specialized knowledge, you know, in each of these different agents all from a single interface, right? And I think that that's sort of the, we're going to start to see this hierarchical, you
know, interaction between agents to do specific jobs that will be really, really a change
in how we interact with things, right?
So like you're saying, I wake up every morning,
I read 20 newspapers or read 20 headlines
or whatever it is, you know,
simultaneously running all of your like morning tasks
and serving you up with the executive summary
of the things that you personally care about,
you know, on any given day,
whether that might be financial opportunities
or news stories or personal ongoings and things and Facebook like catching up on
everything you care about like that is like the simplest use case in the world
of like what this is like is like serving you a daily newsletter letter
for everything that only you fucking care about right exactly you know that
that's that's that's where we that's where we want to and then being able to act on things
In those stories in a streamlined ways, but anyways
I'm I'm I'm just I'm over the moon about it because I think that we're you know, we've been
For me, I felt like we were in a bit of a holding pattern in the AI agent space in general
And after leaving Denver there was a bit, you know,
nobody was talking about MCP yet.
No, literally we've, you and I, you know,
we weren't always together out there, but we, we were at everything.
We basically covered every single event.
And I don't know how many freaking people I talked to.
I don't know, hundreds of people,
not one person even mentioned anything with mcps and even
at different you know talks from people from like near and all these other other people that are
building the ai space no one mentioned it at all so definitely obviously still really early and even
now it's kind of hard if you go try to do research there's not many people talking about it and if
they are they're not in crypto you know like the stuff I found about MCP for the most part, um, they have, they're not doing anything in crypto. They're
just web two people or they're just people that like AI or whatever. And you can find some stuff
like that. Not that I've spent like extensive research, but it seems like most people that
are actually talking about it are in the web three space. And obviously with decentralization,
you gotta be, you can't be decentralized unless you're, you know, in blockchain.
So I think it's going to be super cool to see all this.
And yeah, man, just like let people build and create and explore.
And I know I want to do it.
So like I think it's going to be a lot of fun just creating something that means something to me.
Maybe it doesn't mean something that someone else, but I could share that people like hey you know this is what i like or you know this is what
i'm doing so we'll see it's gonna be fun you know and we're we're all in everybody everybody here in
this call is in crypto so like we all we all care about alpha we you know and we we listen to the
people that we trust and you know maybe listen to them only 50% of the time or less.
Right. But like we listen, we listen to the people we trust, but then we ultimately are responsible for making our own independent decisions.
You know, we deserve the best tools available to have access to the most relevant up to up to real-time information to make a good decision.
That's how we improve or level the playing field
in this day and age.
And so right now, to your point,
all of the, because I don't have the demo
of the product that the, the product
that we're working on right now thesis, right.
Um, until the next couple of days, but, uh, all of the MCP work I've been doing, I'm like
my, my client, I've been like locally hosting these MCP servers and you know, it's, it's
a hard experience.
It's, it's not straightforward.
You know what I mean?
Where people are like, Oh, I can just like pick up this tool and, and do all these crazy experience. It's not straightforward. You know what I mean? Where people are like, oh, I can just pick up this tool and do all these crazy things.
It's not like that.
You have to work for it, you know, for a little while.
And then once everything's sort of like set up correctly, you're like, wow, this is incredible.
It's very powerful.
But, you know, at first it's kind of, it's a bit jarring because you're like, okay, you know, I downloaded Claude and, or Cursor AI and now I want to.
What's your favorite, by the way?
You know, I go back and, I've been going, I've been going back and forth between, between Claude and Cursor.
I, I think that the general consensus is that Cursor is better at coding than Claude.
So that's been, but I think that Claude is better at, from my experience,
has been better at managing the different tooling in each of these MCP servers
than Cursor is.
But that's just my experience in like the the 14 days that
I've been playing with it right so I do think that you know as we're talking
about MCP is a level of standardization memory blocks is an abstraction of that
standardization it allows people to plug and play in a more natural way right so
people don't have to go through all of this like hosting locally like you can have cloud hosted MCP servers and be be swapping them out
and creating your own you know experience right and I think that that's
just again it's something that's just not available right now that will be you
know pretty big game-changer so anyways we've been going for like an hour and ten house. Yeah
The last thing I wanted to ask or mention because we didn't really touch on this
I think it's probably imperative that we do or for all even allowed to yet, but I
I'm assuming at first it may be kind of a beta slash free to use but once it's not I don't want to say
it's not i don't want to say finalized i'm assuming that there's some sort of monetary
Finalized I'm assuming that there's some sort of monetary
structure where people have to either be staking awry or subscribe or like how's that going to work
and then we can wrap it up yeah there's um i think that there's uh i i can't say with absolute
certainty i know that there's a couple of different models being floated. I do know that the early release will be only available for Arise stakers.
I don't remember the exact amount.
Maybe you remember it might have been 300 Arise staked or something minimum
to be able to access the early stage.
But I think the official word will come out of that next week.
And then, you know, in terms of like long-term monetization,
when it really is open up to the public, it's going to be, you know,
there's a chance you can go subscription-based.
There's a chance you can go sort of credits-based where you have to top up
in order to have sort of compute credits.
I'm not sure which direction that is going to go yet.
I think that that's a two question and he will probably need some time on its feet.
But yeah, obviously in the beginning it will be free use.
It will be very closed group.
I think more broadly, though, just in sort of the same way that, you know,
agents have always been a sort of define-your-own-monetary model.
Like, you know, even though we're shifting to see,
shifting perspective on how we interact with agents and what they do
and how to customize them and all
those different things, there's still freedom for whoever is creating one of these different agent
spaces or workspaces, workstations, whatever you want to call them, you know, to define their own
revenue model, right? And so the one that we're sort of leading with thesis that is the sort of DeFi-centric agent space is one model, but that doesn't mean everybody that builds one in the future will have to abide by that same thing at all. and dictate that on their own. So, you know, I think a huge leap, you know,
it's hard to believe that we're, you know, in April,
you know, the beginning of April right now,
and we've been, you know, four months or so on this journey
of this agent ecosystem, Haas, since the, you know,
beginning of December maybe a
little bit earlier than that and just how crazy radically everything has
changed like the whole ecosystem has changed for AI in general it's
completely different today than it was yeah you know four or five weeks ago
it's just like a totally different game you know what i mean so but so i applaud i applaud
the right chain dev team for always like you know we're always early we just gotta dominate this
time that's all like we just gotta straight up dominate because i feel like we're always early
to like every sort of new thing like way earlier too right like i'm sure mcp in five months or six
months will probably be like everyone will be
talking about it maybe even quicker than that but yeah we're we're in first so we just got to
dominate and build the best shit and just be awesome so that's my goal let's make it awesome
you know awesome and we got some we got some events coming up i don't know i don't know what
you know i don't know if i'm gonna go to talking 2049 and do bye at the end of the month. I'm thinking about
I might go if you go, Haas, but
I'll reciprocate
that then. It's like if you're going to go, then I'm
going to go.
Well, you know,
it all just kind of depends
on timing of where we are
in the release cycle because I think one
of the things that we've
talked about a little bit this week
has just been, you know, going forward,
making sure that we have the right partnerships,
you know what I mean?
And we have a standardized way to now,
through these MCP servers,
we sort of have this like standardized way
to plug and play everything,
including like partnerships who have services
or chains or D apps that
are, or data, right.
That are all, that need to be accessible to our, you know, workstations and ecosystem,
right.
There's an opportunity.
So, you know, as that's, hopefully that's, you know, ready for us to go out to Dubai
and just go do business with everybody, you know what I mean?
But if not, then we'll
be up in Toronto.
And regardless,
you know, we will be
on this space.
What are we committing to here on us?
I'm down to do Tuesday.
Tuesday and Friday? This is our first Friday?
Yeah, maybe
I guess, yeah, 11 o'clock
11 o'clock East sounds o'clock East East sounds good
It sounds like a good time. I feel like we need to so once once
Thesis actually launches we need to do like
live sessions where we're
Actually using the product live and like getting prompts and like just like playing with funds
Yeah, I'm not i'm down for that because now now actually you can do you can stream it from like whatever other other social
platforms so you can technically be on youtube or tick or not tick tock um all of this shit right
and it'd be kind of cool to just show people oh yeah we're just messing around and people can come
by and just talk about it or show us what they're working on and stuff like that.
Be more interactive, so to speak.
It would be kind of cool.
Yeah, I think it'll be fun to see people show off prompts that work really well
and, like, found unique opportunities, right?
It's always, like, easy to say by the dip, but, like, you know what I mean?
Finding, like, the unique things that are – alpha that your agent is able to extract because you have a certain perspective.
I think all of those things are going to be really interesting to see.
I'm excited for everybody to get their hands on the product in the next couple of weeks, then we're going to keep it going.
We, you know, have a lot of education coming out, a little bit more specific education
on how MCP works next week.
I think our Tuesday space will probably just go a little bit deeper just into how each
of these tools and resources is called and talk a little bit more about the sort of agentic loop
and what parts of our stack are modular.
And we'll talk about that.
And then towards the end of next week,
hopefully we'll have a couple of live demos
of the product to show that are clean without any errors.
We'll see.
My favorite thing about live demos is that everything always goes wrong.
Yeah, everything does.
Everything does.
I'll have the live plan and then my backup plan.
I'll have some tapes in the can if it's not working.
But anyways,
I'm hyped. Thank you, Haas, for
coming by, hosting.
And we'll talk on
Tuesday next week. Alright, everyone.
Thanks for tuning in. Thanks to whoever listens to
the recording. Have a great rest of your Friday
and a great weekend. Thank you, everyone. Take care.
Adios. Cheers.
