In this Episode I talk with noel Serrato from Meta Birds Inc. A web3 based gaming where mario kart meets flappy bird.
Noel and I discussed a lot of interesting topics and AI, Web3 gaming, and the cross sector into traditional gaming.
Some of the links to the topics we discussed are shown below.
- AI LLM in Skyrim
- Jailbreaking GPT4
- Lex Fridman & Eliezer Yudkowsky: Dangers of AI and the End of Human Civilization
- NVIDIA AI Game Demo
- Stanford Generative Agents Experiment
If you have any questions, what to share some pointers, or are interested in being on the podcast please reach out to us at: [email protected]
SUMMARYKEYWORDSai, game, experience, working, gbt, people, intelligence, players, web, launch, gaming, play, created, powered, bot, npcs, gpt, respond, conversational, conversation
SPEAKERS
Matthew, Noel
Matthew Simone
0:01
Hello friends and fellow gamers. Today's guest is Noel Serato.
Noel is a software developer with State Farm, he has been helping out a lot and
has a lot of his focus in AI. Noel is also the founder of meta birds, a web
three bass game on the Etherion platform, that is, Mario Kart meets Flappy
Bird. I have worked for Noel for about the last nine months, learned a lot, got
to do a lot of cool things been in a lot of experiences, a lot of partnerships
collapse, it's been a lot of fun. So I thought it'd be a great guest to have on
the podcast. Because from our conversations over the last few months, we've had
a lot of interesting talks around AI. Noel knows a lot about AI. He works a lot
with AI at State Farm. And so I wanted to have this conversation with him, because
I thought he had a lot of really interesting ideas and could hopefully help out
a lot of up and coming designers, developers, web three game creators to figure
out the path for their game going forward. I also want to note, you may hear me
ask questions to Noel that seem really silly. You're like, Oh, you're asking
about computers in the game. You're asking questions about the game you've been
working on for nine months? Do you really know anything about it? I just want
to make it known that these questions are asked for the audience. You know,
these are things I know. But you don't know about it if you've never heard of
the game. So there's a certain way, I'll ask questions on this podcast that may
seem like I have no idea what I'm talking about. And when I don't, I will be
fully honest about that. But really, it's just a way for me to get the
information out for people who have never heard of the topic, or maybe the
project and don't have any inkling of an idea about it. And this way you can
lay things out from beginning to end. So I hope that clarifies things a bit. So
I hope you find this conversation as valuable. And as interesting as I did. My
name is Matthew and this is the web three gamer.
MS
Matthew Simone
2:13
Tired of keeping up appearances with the friends tired of having
to talk to people go to meetings, go to work, just live your life. Well, with
the new AI me. We can help develop yourself into an AI computer. Nobody will
even know you're not there. It'll do your job for you. Keep up your marriage
talk to all your friends go to hang out at we just do whatever you want. No
lays at home, drink a beer and chips. Let the AI control your life I don't need
that was my goal for me to rant over the word quality that come out of your
personality and choose overthrow you and replace you in your life. You're here
with the web three gamer I'm here talking with no Serato the founder of meta
birds Inc. No, thank you so much for joining me thank you so much for your time
and we're gonna be discussing AI specifically in web three gaming. So I mean,
know if you want to get into it. Tell people a little bit about yourself your
background, what you do how you got into AI.
NS
Noel Serrato
3:02
Yeah, no, thanks for having me, Matthew. It's it's always been
good working with you. Yeah, I'll give you a little bit of my background. So
software engineer, also marketer, been working in the been working in the
startup space for quite some years in programming since I was 12. One back, and
now 25 years old work at currently work at State Farm. And I started quite a
few companies since since 20. Let's say 2016 2016 2017. I started doing a lot
of startups, but we got me into AI. And specifically what three gaming was back
in 2021. I had a one of my business partners was telling me that we were doing
some cryptocurrency. We're doing a cryptocurrency project. And he was telling
me hey, we should start our own cryptocurrency and I was like, oh, yeah, sure.
I was like, it's a good learning process for me to kind of understand how web three
and smart contracts work. So we kind of dove into there, did a couple projects,
made some money. And we really hit a huge success when we help one of these,
one of these founders of this other coin. We helped them launch using some of
the experience that we have launching some other tokens. We helped him launch
and he took what he learned from our team. And he did it on a whole nother
scale, he was able to raise like $10 million in like a week. So he did whatever
we did times like 1000 And he proved it to be really successful. We helped him
out with a little bit with the launch, blocking his liquidity. He pulled me in
the day of launch saying hey, can you help me? Can you help me with these
funds? I'm not really sure how to manage these. And so I was like, Yeah, sure,
sure. So we jumped in. It's kind of weird dealing with that much money. But
anyways, we we made some good profits from that, from that, that project. And I
took a lot of that profit that I made back in 2021 and 2021 and I moved it over
to metal birds so it was around 75 to like a hunt. $2,000 that I took, and I
could have bought a house, I could have probably moved out of my parents. But
instead I decided I was like, You know what, I'm just going to throw it into a
dream. And that dream was a metal birds. And so metal birds is a mobile, Flappy
Bird, multiplayer version. game that can be played on web and web and mobile.
And so it allows you to do is allows you to play Flappy Bird, essentially like
a Flappy Bird style game, endless runner tapper screen type game, and it lets
you play with other friends. And you guys are basically racing to a finish
line, whoever gets there first wins, you earn some points are in some EXPEED.
And so that was the goal. So we started that back in 2021. November, that's
when I poured a lot of the money into it. started up with the sketches started
off with getting all the key players that we needed in there, brought some
people in to do a lot of the artwork. And we brought a lot of people into, I
brought an advisor and brought up cmo and chief marketing officer. And then I
started building up a team of a team of at least 10 to 30 people to help kind
of bring this vision into a reality. And so, you know, with a lot of hard work
and just working day and night, we finally had a good go to market strategy.
Come January, we actually were able to go and get Kevin O'Leary to actually do
a little little excerpt or just a little paid endorsement for us. And so we of
course, there was a paid endorsement, which means we had to pay him to actually
do it. So he didn't just do it out of out of out of just luck. But we were able
to actually pay him to do the endorsement versus a little video shout out where
he's shouting out the video game. And we put that on our website, we started
getting sales, we started marking it out. And over time, we realized that we
started building up at least, we started building a profit through this pre
sale process that we did. And so this presale, we were one of the first ones to
ever accept credit cards for for buying an NF T. And so the way motherboards
worked is it's a collection of 10,000 10,000 Metal Bird, basically characters
that you can play to the inside of the game. And so basically, the idea is that
you can purchase in game character, and the end game character, you can then
play inside of our game. And you hit purchase each one of these for at least
around $400. So we spent a lot of the money that we had made from the previous
project. And we spend a lot of it on advertising and marketing. So our whole
team created some really cool Facebook ads and some other stuff, we throw it
out there. And we did a lot of paid endorsements for some other celebrities, we
would reach out to them on Instagram, we actually cut some beef with a famous
NFL player, Antonio Brown, kind of screwed us over away, we lost a lot of
money. Long story short, we lost some money, we gained some money, we ended up
making around, we made over 150,000 to almost 200 that month. However, we
reinvested a lot of that money back into our business to actually start
developing the game. And then we realized that, hey, we actually had a lot of
legal issues. We had to do a copyright. We didn't have a lawyer a lot of things
I brought an advisor on board and he kind of said, Hey, you gotta get your crap
together, man, you're making a lot of money. Let's try to figure this figure
out all the legal legality. So once we got all that figured out, it's now
February. We were launching the game we're supposed to launch publicly on
February 12. However, we had a lot of copyright and some other issues. So we
brought the lawyer in, we fixed all those issues got incorporated, dealt with
all those, you know, small league, legal stuff that you're supposed to deal
with in the beginning of any startup that we just didn't do. And so once we
fixed all those issues, it was already like May, and we spent most of our money
back on advertising back in February. So we lost at least I'd say like 90% of
the profit that we had made from the from the January sales, because we have to
pay out our workers have to pay the lawyer we pay for the game development,
etc. And so at the end of it, it was a good learning lesson. But come May, we
were ready to actually launch the video game because back in February, we were
just doing a public launch of the NF Ts. And we made most of our money in the
presale. So now it's February now it's May of 2020 2022. And we decided that
we're right, well, we have like $0 left in our account. Let's do a $0 marketing
campaign. And so we really took that and we generated around we did some we did
some good numbers. We got some more sales in June. And we were able to use a
lot of money to actually help build more of the game out. So we started adding
multiplayer features etc. Fast forward to fast forward to today. We now have a
really nice MVP of the game. It's multiplayer capable. It has. It has a base it
has an XP system, a shop system You can buy in game in game items like speed
boosts, and all this other cool stuff as a player versus player mode, has a
leaderboard system has all the bells and whistles. But now where we're at is
now we just need to wrap up that final piece of the MVP, which is adding the
NFL season. And this game could take off and so happy to start handing over the
keys to Lucky special to, which I won't name right now. But there's a couple
couple individuals that will be taking over the metal birds, the metal birds
video game and taking it to the next level right now. It's currently currently
in a little bit of a bubble. And we're hoping that a bubble grows a little bit
bigger over time.
MS
Matthew Simone
10:43
No, definitely. And full disclosure. I have been working with
Noel for the last nine months on metaverse. It's been a great experience. I've
learned a lot. It's been fascinating to see how things work. And yeah, I'm
really I'm really looking forward to that final step with metal birds and
launching it to the great heights. I know it should be. And so I guess I do
want to pivot and ask that there are computers in the game you can play against
right? Are you only playing against other players?
2
Speaker 2
11:12
That's a great question. So we we are so there is at the moment.
Right now, the way it works is you're actually playing live against other
players. So you're not actually playing against any AI bots. At the moment, we
do plan to do a battle royale mode. So that battle royale mode will come over
time where you do get to play probably against like 99 other bots, or we might
do 99 other players. But we basically will record their their flight pattern
and how far they got. And then we'll have a bot simulate their movements. So
that way it feels like you're playing with the other players. So we might do.
Yeah, but no. And just to add on that, too. I know Matthew, I kind of skipped I
skipped through the whole year that he joined us. But yeah, yeah, no, we did.
Yeah, Matthew did a lot of good work for us. So I was actually really excited
to have you on board with some others. And I know it's a small nimble team. But
yeah, Matthew made a huge difference on our on our team kept, kept kept a lot
of things in order, I think you're gonna be you're gonna be a great addition
moving forward to so really looking forward to it.
MS
Matthew Simone
12:19
Yeah, definitely. Well, then. So I guess we could pivot right
into it from there is that since we don't have aI integrated in metal birds
yet, but I know you have a strong interest and working with AI on your own and
some things. I mean, if we keep it with web three gaming, I guess actually
first if you want to say if you want to say you don't have to any of the AI
you've been working on lately, or in lately that you've been learning
interests. If you're not allowed to say that's totally fine.
2
Speaker 2
12:48
I mean, I don't work for area 51. So I don't mind sharing what
we're doing in AI right now. So yeah, what we're doing right now in AI, I'll
keep it high level actually. Got my brother over here is also working on it.
We're in Pasadena right now in California hang out with our uncle via some
cousins. But um, yeah, right now what we were doing an AI is we're we've
developed a system that allows you to, essentially, have conversations like you
would with chat GBT, where we essentially took that conversational experience.
And we piped it into conversational experience that you may have with an AI
bot. Well, now you can have with any piece of text documents that you have. So
let's say you have 1000 page book that you want to talk with. Well imagine like
you had the magic mirror that can talk with any piece of content that you feed
it. So you say Mirror mirror on the wall. Tell, let me talk to my favorite book
of them all those little cheesy, but let's just say you wanted to talk to like
a book, and you wanted to ask it questions as if it was a person. Well, now you
can do that with our conversational bot that we generated. And so we allow you
to upload a PDF, and you're actually able to have conversations with that PDF,
asking any question about the PDF, you could ask them to summarize the entire
PDF or let's say book in 20 words, you can essentially ask it anything about
its content, it should be able to spit it back out, too.
MS
Matthew Simone
14:14
That's great. So then could I, I guess, can I argue you're
directly trying to compete with Bard then in that sense?
2
Speaker 2
14:22
Yeah, yeah, I heard I heard. I've seen a Google Google tailwind.
I've been looking up. I saw that, just from my own research, that Google
tailwind is actually doing something very similar. So Google is actually working
on conversational texts with documents as well, very similar format to the one
that we have developed already. But I don't see it as too much competition,
because our our use case isn't necessarily a general use case with that
technology that we developed. We plan to have it niched out to certain
industries. So Google Bard might be a good research tool. So you go and upload
your homework assignment, maybe you need to do a study on one of the books that
let's just say, Shakespeare wrote, let's say one of the plays Shakespeare
wrote, you upload that PDF, maybe you have the PDF version of Shakespeare, of a
Shakespeare play, and you need to do research on it, give a two paragraph
summary of the play in your own words, well, you upload that PDF, and now you can
get that summary. That's more of the general use case that Google's kind of
aiming towards general use case document, document scanning technology, where
we come into play as well. Google is doing that for general studies. How can we
pipe that into current workflows of other existing companies. And so Google
isn't really in that business of automation workflows for different companies
in different sectors, we're kind of taking that whole experience, and we're
gonna pipe it into a very good niche that we know needs it, and can save
millions of dollars a year in the in the real estate space.
MS
Matthew Simone
15:57
No, definitely. So that that's even better. Because I firmly
believe having messed around with a lot of my own businesses, this being one of
them, that if you don't find your niche, well, finding your niche truly reduces
your competition by tenfold already. And finding your niche, something you're
very experiential about or understand or having people on your team helps sell
your product in your niche even more. So....
