June 24, 2025 • 45 mins
In this episode, I’m joined by Erik Faz Jurgensen, a visionary developer and one of the most creative builders I’ve met. Erik has worked across Web3, government tech, and AI innovation, often managing more than 20 projects at a time. He taught himself to code using YouTube and hands-on experimentation.
This conversation explores what it means to “vibe code,” why Erik avoids full YOLO-mode with agents, and how he built an AI assistant that manages a whole team of task-specific AI agents. If you’ve ever wondered how to go from gamer to full-stack AI-enhanced dev, this one is for you.
🔥 Topics We Cover:
- How Erik taught himself to code without a CS degree
- His journey from gamer to solo dev with a profitable freelance business
- Vibe coding: what it means, why it works, and how AI accelerates everything
- How Erik built an AI agent team using Relevance AI and n8n
- Why developers should say “yes first, learn later” in the AI age
- What people are still sleeping on when it comes to AI adoption
🧰 Tools Mentioned in This Episode:
- Learning: YouTube, Stack Overflow, ChatGPT
- Coding + IDE: Cursor, GitHub
- Voice + Media: 11 Labs, Leonardo AI
- Models: GPT-4, Claude, Gemini (Google X), xAI (Twitter)
- Low/No-Code Platforms: n8n, Relevance AI
- UI Tools: Bolt.new, Lovable.dev, v0.dev
📚 Mentioned Book:
Who Not How by Dan Sullivan and Dr. Benjamin Hardy
🎥 Tutorial Discussed:
Relevance AI: Build an AI Agent Team (YouTube)
🔗 Connect with Erik:
- Portfolio: https://portfolio-erik.vercel.app/
- Twitter/X: @erikfazj
Want to be featured?
Have a compelling AI use case or experimental workflow? Apply to be a guest at howiaipodcast.com.
Enjoy dining out? Use my inKind link to get $25 off your next $50+ meal. This is a referral link and I may receive a small reward if you use it.
More About Brooke:
Instagram: thebrookegram
Website: brookex.com
LinkedIn: Brooke Gramer
More About the Podcast:
Instagram: howiai.podcast
Website: howiaipodcast.com
"How I AI" is a concept and podcast series created and produced by Brooke Gramer of EmpowerFlow Strategies LLC. All rights reserved.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Brooke (00:02):
Welcome to How I AI the podcast featuring real people,
real stories, and real AI inaction.
I'm Brooke Gramer your host andguide on this journey into the
real world impact of artificialintelligence.
For over 15 years, I've workedin creative marketing, events,
and business strategy wearingall the hats.
(00:22):
I know the struggle of trying toscale and manage all things
without burning out, but here'sthe game changer, ai.
This isn't just a podcast, How IAI is a community.
A space where curious minds likeyou can come together, share
ideas, and I'll also be bringingyou exclusive discounts, free
trials and insider resources soyou can test drive the latest
(00:45):
tools and tech yourself.
Because AI isn't just a trend,it's a shift.
The sooner we embrace it, themore freedom, creativity, and
opportunities we'll unlock.
How I AI is brought to you inpartnership with The Collective
designed to accelerate yourlearning and AI adoption.
I joined the collective and it'scompletely catapulted my
(01:07):
learning, expanded my network,and show me what's possible with
ai.
Whether you're just starting outor looking to refine your AI
strategy, the collective givesyou the resources to grow.
Stay tuned to learn more at the end of this episode, or
check the show notes for myexclusive invite link..
Hello everyone.
Welcome to another episode ofHow I Ai.
(01:29):
I'm Brooke Gramer, your hostToday I have a very special
guest.
His name is Erik Faz Jurgensen,and he's a builder, he's a
visionary mind, and he's thefirst developer I've had on my
podcast, so I'm excited to divedeep into his journey and his
background.
Erik, welcome.
Thank you so much for beinghere.
Eric (01:49):
Thank you for having me, Brooke.
Brooke (01:51):
Yes, and I love to leave the floor open for you to share
more about yourself.
Eric (01:56):
Okay, so I'm a software developer.
I've been coding for the pastsix years almost.
Mm-hmm.
I have my own business calledSpy Dev, which is a one man.
I'm a serial freelancer.
Brooke (02:08):
Okay.
Eric (02:08):
I just have an a brand name behind it, but yeah I just
code all day.
I love coding.
I was a huge gamer before Istarted coding.
And then I saw YouTube video onhow to gamify your life and get
dopamine hits out of simplestuff the same way you do in
video games.
So coding I literally haven'ttouched a game controller in
(02:31):
five years.
Brooke (02:32):
That's so fun.
One YouTube video changed yourwhole trajectory.
Eric (02:37):
Yes.
Brooke (02:38):
My brother's a gamer.
He loves that space as well.
He's also very smart.
He's a background in engineeringand works in cybersecurity.
So share more about that journeyof what was like the first
initial step and what did youstudy in school exactly and how
did you really get into thisspace?
Eric (02:56):
So in school I was studying, innovation and
development engineering.
And I actually took some codingclasses and I hated it.
Oh.
I thought I was gonna likecoding.
I really, I hated coding and Iwould cheat on my exams to pass
coding classes, and then Istarted making websites.
And started running my solobusiness and eventually dropped
(03:17):
out from college.
And then a friend of mine whohas a big company in Mexico,
they raised money and he waslike, Hey, I want you to join my
team.
I want you to code with us.
And I'm like I don't code, Idon't know how to code.
And he's like, that's all right.
We'll teach you.
Wow.
So I accepted the gig and theirway of teaching me was just,
okay, these are the programminglanguages.
(03:40):
This is what you have to do.
Look it up.
So I started looking things up,like handing something in, and
then they were like, okay, nowdo this.
Now do this.
And I started having fun and Istarted coding that way.
And since then I've just workedon a lot of different crypto
projects.
(04:00):
Mainly crypto projects, codingin Web3.
Crypto changed my life forever.
The space, I didn't knowanything about it until I
started coding inside of it.
And yeah, been part of hugeprojects in crypto and now of
course AI is part of my, like Ido AI projects, but mostly I use
AI all the time.
Brooke (04:21):
How many projects do you have at any given time?
Eric (04:24):
So the most I've had in any given time is 23.
Brooke (04:27):
23,
Eric (04:28):
23 websites.
Wow.
Right now, I guess it's five orsix.
Okay.
Brooke (04:33):
Yeah.
Do you primarily work onwebsites or what kind of
projects do you feel like youspecialize in?
Eric (04:39):
Before it was websites.
Now I actually don't like makingwebsites that much.
I like to do like webapplications and more robust
systems.
The best project I'm working onright now, the one I enjoy the
most is called Veeam.
It's like my full-time job.
Okay.
And we're basically creating a,an AI platform where people can
generate video memes with ai.
(05:01):
So we're training some modelsand we're getting a lot of
context from clips, from memesand using that to create video
memes.
Yeah,
Brooke (05:11):
I love how.
You have fun with it.
And that's how you startedlearning was from a place of
just excitement.
Eric (05:18):
It's excitement.
And also I truly believe ingetting paid to learn.
'cause that's how I started.
So now whenever a new clientcomes in and says Hey, I wanna
build this platform, even if Ihave no idea how I'm gonna do
it.
I'll just say yes and learn.
So by the end of that project Ilearned a lot.
I got paid.
Mm-hmm.
I had fun.
So.
(05:39):
It's cool.
Brooke (05:40):
So do you ever take on projects where you don't know
how you're going to solve theirproblem or complete the project?
Eric (05:47):
All the time.
Brooke (05:47):
All the time.
All the
Eric (05:48):
time.
Brooke (05:49):
And how do you learn, how do you figure that out?
Eric (05:51):
YouTube.
YouTube.
YouTube.
Yeah.
Before it was YouTube I creditYouTube.
As my reason for dropping out.
Brooke (06:01):
Wow.
I call
Eric (06:01):
it YouTube University.
'cause literally YouTube haseverything, not only in coding,
whatever you wanna learn.
Yeah.
Changing the tire to coding.
It's all on YouTube and thenfirst it was YouTube, then GPT.
Brooke (06:15):
ChatGPT is how you learn everything now.
You just ask it what you need toknow.
Eric (06:19):
Now I don't learn.
Now it does the work for me.
You know?
Brooke (06:22):
It does the work for you.
Yeah.
Most of
Eric (06:23):
the work.
Yeah.
Brooke (06:24):
That goes to my next question is what are you using
right now?
What's your technology stack?
What are your favorite?
I.
Tools or do you create your ownin-house for everything?
Eric (06:34):
So my, my favorite tool, hands down is this app called
Cursor.
Cursor, which is what we call anIDE.
I don't know what it stands for,but basically it's like my
coding tool where I write codeand it was the first of its kind
that integrated ai.
So now I have a sidebar uh, withAI that has all of my project as
context.
So if I tell it, Hey, now createthis new tab, or this new page
(06:57):
for the project, or whatever, itknows that design I'm using the
colors fonts, styles, everythingI already have in the project.
So it'll do a great job.
So I mainly use cursor.
I pay for GPT for my sisters touse it.
Mm-hmm.
Uh, I cursor is my everyday app.
Everything app.
I just code all day and I justpay for different subscriptions
(07:18):
to try out 11 labs, Leonardodifferent apps, just to like a
new thing comes up, I'll pay forit, test it, forget I
subscribed, pay it a couple moremonths and then unsubscribe.
Brooke (07:30):
That was my next question is how much do you
think you're spending a month onai?
Eric (07:35):
So thankfully my projects pay for my stack.
Okay.
Most of my stack, yeah.
I just pay for like differentsubscriptions I wanna have fun
with.
But I'd say probably a hundredbucks a month.
Brooke (07:46):
Okay.
Eric (07:46):
Which is not that much.
Brooke (07:47):
No, not at all.
And it
Eric (07:48):
could actually be$20.
And I could do the same workwith 20.
'Cause cursor, what's greatabout Cursor is it has ai but
it's not specific to one model.
So inside of Cursor, I can useGPT Claude from Anthropic,
Gemini, from Google X, AI fromTwitter or X.
So I got all in one for 20 bucksa month.
(08:09):
Wow.
And.
It's crazy.
Brooke (08:12):
That's cool.
So what else are you excitedabout learning right now?
What is new to you in thisspace?
Eric (08:19):
What's exciting and new for me is catching up, right?
Catching up.
Catching up.
Like I, I feel I'm superadvanced, but at the same time,
being at the forefront ofeverything that's happening
gives me like anxiety.
'cause you never you don't knowwhat is better than what, right?
Last week, Claude four wasreleased, and then today GPT 03
Pro was released.
(08:39):
And they made 03 super cheap fordevelopers.
But then tomorrow, GoogleGemini's newest model work will
come out and they're all reallygood.
The hard part and the excitingpart at the same time is
figuring out which one best fitsmy needs in coding.
Brooke (08:58):
Do you feel like there's a lot of.
Overlap or they're still uniquein what it is that they do for
you?
Eric (09:04):
For me, I think there's a lot of overlap.
A lot of overlap.
I just go through Twitter andsee what other coders are saying
about the models and okay, nowI'll use Claude.
And if I start seeing, if myalgorithm starts saying oh no,
Gemini is now the best.
I'll just switch to Gemini.
But realistically all of them.
Do a great job.
And it's crazy, like workingwith them every single day
(09:27):
actually makes me notice howgood and how better they're
getting over time.
Literally by the day.
Yeah.
Like yesterday, there's a bug Icouldn't solve with Claude
today, it'll solve it, yeah.
They just keep getting betterall the time.
It's scary.
Brooke (09:41):
Do you have a community of coders and developers that
you're in?
Nah.
Do you feel like you just chatwith people on like Twitter?
I
Eric (09:48):
just, yeah.
Yeah, I just chat with people onTwitter.
Brooke (09:52):
Yeah.
Eric (09:52):
Yeah.
Because before there was thisthing called Stack Overflow
where whenever you had aquestion or something you
couldn't fix by yourself, youwould post a question post your
code, and a community of coderswould help you.
Now, ai.
It is my community.
I think I, I talk to the robots.
Brooke (10:10):
Yes.
We're in a community together,the collective, and we just had
a
Eric (10:14):
Mastermind.
I, we're not all coders, that'swhy I didn't say that.
We're not coders.
Yes, it's not a coder community,
Brooke (10:18):
but it's more of an AI in general community.
We recently had a mastermind andHacker House and you're
absolutely the most favoriteperson there because what does
everyone need to have theirdreams come true?
Developer, someone who knowscode.
So I'm curious because everybodyhas an app idea.
Everybody has a business idea.
(10:39):
Everybody has a super uniqueproblem that they're solving in
their personal lives, and thenext step is always how do I
make this come to life?
How do I.
Build.
And the next step is obviouslyto work with a developer or have
a CTO in your company.
How does someone go aboutprocuring that?
(11:00):
What would be the best way?
And speaking from a developer,how do you like to work with
people?
How do you approach thatrelationship?
Eric (11:08):
So I really like working freelance based.
Not getting a full-time job Yes.
Or something that will make meonly work in that project.
Brooke (11:15):
Okay.
Eric (11:15):
cause now I think it's a win-win.
'cause now we devs can workfaster.
Way faster.
You don't have to hire a team offive devs.
Mm-hmm.
You can have one or two devs.
And if you allow them to work inother projects we'll deliver
faster, right?
We'll do your project, we'll doit fast so we can move on to the
next project.
Brooke (11:32):
Okay.
Eric (11:33):
So I really like that.
And from the employer's point ofview I think it's super
important right now to focus onsearching for devs or people
that.
That are hungry to stay up todate.
Mm-hmm.
Right.
So before you needed seniordevelopers mm-hmm.
with 15 years of experiencestuff like that.
Right now, it's funny'causethose same guys are the ones
(11:55):
with the most resistance to ai.
They're so pro, they have theirown way of doing things.
And they're super expensive.
And they all work in Facebook,apple, Netflix, Google.
Yes.
Now like junior developers orpeople who haven't even
graduated can get way more workdone with way better quality.
I believe, than senior devs.
Brooke (12:15):
Wow.
Eric (12:16):
So that shift is crazy.
And I'd say just look for peoplethat, that wanna stay up to
date.
'cause if you're staying up todate, the tools do the most part
of the work mm-hmm.
right now.
And it doesn't have to be superexpensive.
You can get your MVPs or ideasbrought to life super fast,
super, super fast.
Wow.
And allow the devs to work ondifferent projects.
Brooke (12:37):
Yeah.
That's awesome.
One thing that we chatted abouta bit last week was vibe coding.
If you could share in your ownwords what it is and what you
see the future of Vibe Codingand ai.
Eric (12:50):
Yeah, so Vibe coding is what we now do every day as
coders.
So instead of writing code,we'll type in natural language
like, Hey, please add threecards, or Please change the
background color to red, orwhatever.
AI will do the job.
So that's why it's called Vibecoding.
And then you just review.
You have to do minor changes,like super small changes.
(13:11):
'cause the models are supergood.
But life coding is basicallyinstructing ai on what you need
or what needs to get done.
Yeah.
And then just overseeing that.
And there's a ton of apps.
So inside of Cursor, what we dothere, there's a lot of browser
apps like Bolt.new Yes.
Lovable.dev.
Yeah.
v0.dev from Vercel.
(13:32):
Those are all great, but I thinkthose are great for beginners.
Because you just type your idea,it'll bring it to life.
But once you go into the codingTrenches Cursor does the same
thing.
Whatever you can do in thoseapps, you can do in Cursor, it
just looks a bit moreoverwhelming'cause you do need
to know more stuff.
But it's super cool'cause youcan change the models and.
Let's say I ask GPT to fixsomething and it gets it wrong,
(13:55):
so I can revert and now tellGemini the same prompt, until
one of them gets it right.
And then I'll just stick tothat.
Yeah, that's pipe coding.
Brooke (14:02):
I wanna touch a little bit more about the anxiety of
being into this space.
I thought that was like veryvulnerable of you to share just
'cause on from the outside wewould think that developers feel
very confident and they're veryexcited and just sharing how, oh
my gosh, there's so much to stayon top of, and feeling like, you
know, all the right things.
(14:24):
Do you see that going away, ordo you think it's gonna just
always be this rat race oftrying to stay on top of
everything?
Because
Eric (14:32):
I think it'll get harder actually.
Yeah.
So what's crazy is I was blessedto use Chat GPT the day it came
out.
Wow.
I was scrolling through Twitter,saw this new tech.
Started using it, not even forcoding, just asking stupid
questions.
Yeah.
And I was like, wow, what isthis?
And then we started using it forcoding.
And we were like, wow what isthis?
(14:53):
But when that came out I don'tremember the exact roles or the
exact order, but basically wethought developers and
designers, out of all the jobsin the world, were gonna be.
The ones taken last by ai.
And it turned out to be theother way around.
AI first came for designers.
You can create images, coolvideos, cool websites all
through prompts.
(15:13):
And then it's coming for usdevs.
Mm-hmm.
Because now like vibe coding,people who have never coded in
their lives can do apps.
Can do websites in minutesbefore it took weeks, maybe
months if it's an app.
And also two or three years ago,computer science was the most on
demand skill.
(15:34):
Yes.
Two or three days ago, news cameout that it's the number one
unemployed wow.
University degree right now inthe us, like from being the
number one employed to the leastemployed.
Wow.
It like computer science is nowworse than arts and history.
Wow.
(15:54):
Which is crazy.
Which is crazy.
That's part of the anxiety,right?
You've gotta either keep up or,though AI won't take your job,
but someone who uses AI will.
That's uh, saying I'll neverforget when I read it.
So it's anxiety, it's fun.
'cause if you stay up to date,it just keeps getting better.
But for people who refuse to usethe tools to try new things out
(16:15):
they're gonna get wiped.
Brooke (16:17):
For instance, someone who vibe codes but has a
computer science degree is gonnabe able to do it way better.
Yeah.
Than someone who doesn't.
So I still think it's importantto have this background and
skill and knowledge, whetherit's YouTube University or real
university.
Eric (16:31):
Yeah.
But probably for a year.
Brooke (16:33):
Just a year.
Eric (16:34):
I think maybe less, like right now, yeah.
You do need a dev.
Let's say you vibe codesomething in Bolt and you wanna
add your specific domain to it.
You probably, you can YouTube itor you can have a dev do it.
In the future, all of thesetools are getting way better.
Like before it would only do thedesign.
Now it does the code.
Now you can integrate Stripe forpayments and you can integrate a
(16:54):
database.
Before it was just the frontend.
So now you can deploy directlyto GitHub, which is the Google
drive for devs.
So every week all of these toolskeep getting better and I feel
that gap from having to have acomputer science degree and not
having it.
It's shrinking so fast.
Brooke (17:13):
So it's valid to not be going to school for that.
Eric (17:17):
Yeah.
Brooke (17:18):
Wow.
Yeah.
So what, I think
Eric (17:19):
it's valid not going to school for anything, to be
honest.
Brooke (17:23):
Yeah.
What do you think, what do youthink people should be learning?
What do we need to know?
Eric (17:26):
I think people should learn how to learn.
Brooke (17:28):
Learn how to
Eric (17:29):
learn how to learn.
Whatever it is you wanna learn.
Yeah.
Just have the passion and thediscipline of Google it, you
know?
Mm-hmm.
or now Chaskt it that was SamAltman's way of Google it for
ChatGPT Mm-hmm.
Chaskt it
Brooke (17:42):
Chaskt I haven't heard that one.
Eric (17:44):
Yeah.
Chaskt it YouTube, it, Googleit.
All the infos out there.
And i, don't like universitiesbecause there's no way they
would be teaching everythingwe're learning right now.
Yeah.
For example, everything wetalked about in the hacker
house.
Yeah.
Not only us, like everything youcan find on YouTube, I feel
universities have their boomerway of teaching.
And super methodological, likestep by step and using AI is
(18:08):
cheating.
And they make you study on yourown.
Do homework on your own, likeyou can collaborate'cause that's
cheating.
Tests are on your own.
But then you go to the realworld and it's just teamwork
everywhere.
So I need to encourage everyoneto drop out.
Actually, I need to do it.
'Cause everything you wannalearn, you can learn it way
(18:29):
faster, way better outside ofschool, in my opinion.
Brooke (18:33):
Do you work in a team?
Eric (18:34):
I work in several teams.
Yeah.
Brooke (18:37):
That's awesome that you have that support and you made
such a valid point.
We only rarely do team projectsin school where it's like the
opposite.
It's mostly teams and then it's,
Eric (18:48):
you need a team.
You need a team in the realworld.
Like I, I am a solopreneur.
But I think, what's that sayingthat says you can go faster by
yourself mm-hmm.
but further with a team.
So I do think it's all aboutteamwork.
Funny enough, I think AI can bepart of that team, if not most
of the team.
(19:09):
Your
Brooke (19:09):
co-pilot.
Eric (19:10):
Your co-pilot.
Yeah.
And co-pilots.
Co-agents.
New Tech is coming out forbasically everything.
Um.
And it's just weird'cause I'mreading a book called Who Not
How, and it talks about theimportance of having teams.
Yes.
But this was written, I don'tknow how many years ago, and I
just keep asking myself okay,valid.
It's true.
You have to set bigger goalsthat you can't achieve by
(19:30):
yourself.
But how many of those who's cannow be different AICUs?
Because you can literally use itfor everything.
I use it for code, but you cando whatever.
You can run ads, you can dodesign, social media content.
Yeah, automate everything.
I don't know.
It's scary, but fun.
Brooke (19:45):
What does your AI copilot team look like?
Do you have a bunch of agentsworking for you at one time?
Eric (19:51):
Yeah.
Yeah.
So my team would be, I have mypersonal manager who's ai.
Okay.
He's the only agent that talksto me.
Okay.
I only talk to him and based onwhat I ask it or what I say that
needs to be done, it'll talk toa team of AI agents and delegate
those tasks to each agent.
'cause in the AI world, it'ssuper important to have,
(20:11):
specialized agents.
Not having one charge GPT thatsays, ask me anything.
And it'll do everything.
So if you train each agent to bevery good at emails, very good
at ads, at coding, whatever, andyou have one personal assistant
that is in charge of delegatingthe tasks to them that's my
team.
Brooke (20:29):
And how did you build out your team?
Eric (20:32):
With a lot of tools.
I love N eight N.
Okay.
We went over that in the hackerhouse.
Yes.
I forgot to mention Cursor.
N eight N is super great.
It allows you to not only buildagents, but create automations
with those agents.
So I've built a team with Neight N also an app called
Relevance ai.
Which is basically for that tobuild your AI team.
(20:52):
And actually the idea of.
Having a personal assistant andthen delegates tasks to other AI
team members came out from aYouTube tutorial.
Wow.
On relevance ai,
Brooke (21:03):
do you still have that YouTube tutorial?
Of course.
Maybe I'll link it in the shownotes.
Eric (21:07):
Yeah,
Brooke (21:07):
that's a really good one.
And so when you wanna talk toyour agent manager, how do you
talk to it?
Eric (21:12):
You can send emails, you can send it a WhatsApp message.
You can do telegram.
I like Telegram better.
'cause WhatsApp like Facebook orMeta has a lot of permission
issues.
And you have to have a businessand verify the business and get
approved and stuff.
Telegram, you can just connectit or you can even connect one
or several agents to Slack.
(21:32):
So you can send a Slack messageto your personal assistant and
then have it delegate the tasks.
So it can be the trigger, as wecall it can be anything.
You can imagine something inyour CRM an Instagram message,
whatever.
Okay.
Brooke (21:48):
Cool.
And so it just reports back toyou when the project's finished
or it needs additional support?
Eric (21:53):
Yeah.
You can either enable YOLO mode,which is you don't need any
approval from me, just get thetasks done and post or do
whatever you have to do.
Yeah.
Or you can have approval, right?
You can have approval by thepersonal assistant, or you can
also integrate to your team.
Agents that verify that the workis being done correctly, right?
So the agents will talk betweenthemselves make edits or
(22:17):
whatever needs to be done.
And then post, I still likeapproval.
I think it's important'cause younever know Yes.
When the agents can hallucinateand just start doing stuff on
their own, which can be reallydangerous.
You can approve, you can get anemail report or a message or
whatever that just says, this isdone, this is the post.
Do you like it?
Yeah.
Boom.
And it'll post.
Brooke (22:37):
I've heard that there's a workaround for hallucination
using the RAG system.
Do you use that?
Eric (22:43):
I do use it, but I I don't think it solves hallucination.
What RAG does is basically, it'sa better way of giving your
agents knowledge.
Brooke (22:52):
So
Eric (22:52):
They will hallucinate less on the knowledge they're trained
on, but that doesn't meanthey'll hallucinate less on the
actions they're taking.
For example, we built a reallycool agent and he had replied to
new emails that were coming.
Basically, it replaced perfectlyusing rag, but it started
replaying two emails from threeyears ago.
Brooke (23:10):
Oh no.
Eric (23:11):
And stuff like that.
So RAG helps for, from aknowledge standpoint, but not
like action.
Brooke (23:18):
Yes.
I think if I haven't made agentsyet, but I feel like I would
like to have the approvalprocess.
I wouldn't go YOLO mode.
Eric (23:24):
Yeah.
I don't do YOLO mode even incoding.
Okay.
I always have to approve evenwhat sources is the agent gonna
search?
Yeah.
Or what, and I mean, it's bittedious, having to click, accept
or approve or whatever.
Mm-hmm.
But it just makes it better.
You lose, you use less tokens.
It's less expensive.
Okay.
And it gets it right more thanif you just do YOLO mode right
(23:47):
now, I think in a few months youcan just have your mode on
everything.
Brooke (23:53):
Love dining out.
Here's a little gift for you.
I've been using InKind and it'shonestly a game changer for food
lovers.
It's like a universal diningwallet that earns you 20% cash
back at amazing restaurants,cafes, and bars nationwide.
And as I thank you for being alistener, you can grab$25 off
your next$50 tab when you signup with my link in the show
(24:16):
notes.
Just pay with InKind, earnrewards and use them on your
next outing.
A quick heads up.
This offer is for new InKindusers and covers food, drinks
and tax, but no tips or fees.
The credit expires 30 days afteryou claim it, so don't let it go
to waste.
Check the show notes for myaffiliate link and enjoy your
(24:38):
next meal on me.
Well kind Of.
So with everything that you knowin this space and its
adaptability and its ease ofuse, what do you think a lot of
people are sleeping on right nowwhen it comes to the benefits of
ai?
Because even I feel like.
I'm only tapping into 2% of whatI could be doing.
(25:00):
What do you think is somethingthat people could be doing
beyond ChatGPT as a beginner,what would you recommend they
could be immediately using forai?
Eric (25:11):
I think what people sleep on the most is, realizing they
can use AI for literallyeverything.
Brooke (25:16):
Yeah,
Eric (25:16):
literally everything.
What do you wanna do?
For example, for the podcast,you wanna run ads for the
podcast, you wanna reach out topeople.
Like anything you can imaginecan be done with ai.
Mm-hmm.
But I feel there's a lot ofresistance in my personal
experience.
I've seen a lot of resistance onpeople saying if I start using
AI for this and this, then I'mgonna stop using my brain.
And I'm gonna become obsolete.
I'm gonna become lazy.
(25:38):
Yeah.
That's one common denominatorI've heard in everywhere.
Like here in Bali or in the USor in Mexico.
They're like I don't wanna stopusing my brain.
Yeah.
And it is just super funny'causeI don't feel like it's, you're
not stopping your brain, you'reactually automating repetitive
tasks and liberating your brainto think about new stuff.
(25:59):
Yeah I just question everything,right?
Question everything, everythingyou're doing, whether you like
it or you don't okay.
How could AI do it or how couldAI help me do it better?
I think it's a mind shift that'shard to get to, but I think it's
necessary'cause tech's moving sofast, right?
Yes.
And I feel like the tech that'sout there right now is just the
(26:21):
tip of the iceberg.
And they have new tech ready,but they haven't released it.
'cause they first want people tostart using GPT for most things.
And start learning how the worldis gonna work before they just
automate your life, wow.
So I think stop havingresistance and question
everything.
Question everything.
And for people who have thisresistance I was talking to a
(26:42):
friend and my friend was sayinghow ai, artificial intelligence,
what's artificial is not theintelligence artificial are the
tools we're using to talk tothis intelligence.
'cause this intelligence hasbeen out there.
He compares it to Mother Nature,right?
There's intelligence everywhere.
In plants.
In trees, yes.
Like everywhere.
And we're just creating thetools to tap into that
(27:06):
collective intelligence.
So I think it's inevitable andpeople should, instead of being
scared learn to have fun withit.
Brooke (27:13):
That's really good advice.
Do you feel like that's how youapproached it in the very
beginning?
A
Eric (27:18):
hundred percent.
Yeah.
I've been so excited ever sincesince day one, super excited.
And I, and even I I use it forcoding, but I feel I should use
it for way more stuff likerunning ads.
I've never run ads in my life.
Never.
And I need ads to grow mybusiness or whatever I want do.
Yeah.
I should start using it evenmore.
Brooke (27:37):
You can clip up this podcast and run some ADs
Eric (27:40):
I should.
Brooke (27:41):
I should.
I should.
Eric (27:43):
'Cause I feel like once I do it and I start getting
results, it'll be like, wow.
'Cause three years ago since Iwanted to run ads, I would never
do it.
'cause I had to learn ads andlearn AB testing and learn have
someone that creates the copy,have someone that generates the
image.
Someone that's a master like admanager.
Now a, I know ai.
(28:04):
Can't do this, and I'm saying itso I hold myself accountable.
That I need to run.
At least thats, there's a lot ofstuff I could use AI for Yes.
That I should use AI for, but atleast running ads or for
example, I hate being in frontof cameras.
And I really wanna start myYouTube channel.
And right now it's so easy toclone yourself.
And create all of thesetutorials, right?
I don't even have to be in thecamera and I'm still putting it
(28:26):
off and I know it's easy.
I've cloned myself.
I just gotta, I don't even haveto write the scripts.
I just gotta get started.
So another piece of advice to meand to everyone is just get
started.
Whatever your idea is, yourproject.
You just gotta get started.
And I think you're a really goodexample.
I've been talking about thispodcast with the guys and it's
(28:47):
crazy how you just said, youknow what, I'm gonna start a
podcast.
And now I don't know whichnumber of guests I am, but you
got started, right?
Yeah.
You got started.
And that's super important.
'cause now I feel creativity isdemocratized.
Like now.
Now there's no.
Doing whatever is not hard.
(29:08):
You just gotta get started.
I just gotta get started.
Brooke (29:10):
Just get started.
Yes.
This past week together with allof us coming together for the
conference we were in, reallyinspired me to do more with the
podcast and be doing morepromotions and getting out there
more.
So I'm excited to just beputting more energy into it and
I don't know, I think you'relike, guest 15 by now.
(29:31):
That's, yeah.
It's cool.
That's fish.
Eric (29:34):
Super.
No, congrats.
That's crazy.
Yeah.
Brooke (29:36):
Thank you so much.
So you spoke a lot about thebenefits.
I'm curious, have there been anydownfalls or Oh crap moments
with using ai?
I think it's fun to hear likehow we've had learning moments
with using ai.
Eric (29:51):
Yeah, I feel coding projects has become so easy
using AI that we almost take itfor granted.
Yeah.
And then when we actually pushthe code to production, to live
mode.
We don't realize, like we don'teven double check now.
We're so confident.
It's doing a great job.
Sometimes we don't even check.
And let's say it fixed a hugeproblem.
(30:12):
So now that problem's fixed, wepushed that, but we didn't
realize that fixing thatproblem, it messed five things
up.
Oh, or the videos are notworking or whatever.
Oh.
And it has happened more thanonce.
Especially in a team.
'cause it's not only me, it'sthree of us.
Let's say we are all using ai.
I don't know what these otherguys are coding, for example.
(30:33):
So I might use AI to fixsomething missing up their work
without ever realizing,'cause Inever touched their code.
So yeah, I think that's whyhaving YOLO mode is a huge
learning moment.
But I think that's the only, oh,crap moment.
I fucked things up for the code.
You can always revert, right?
But yes, sometimes it's not usthat realize.
(30:54):
It's clients sending messages,like horrible user experience in
this part, and we're like, oh,shit.
That wasn't a problem yesterday.
That, that's the only,
Brooke (31:02):
yeah.
So nothing too bad.
My next fun question for you is,since you're in that space of
creating so much as a developer,what are you wanting to wave a
magic wand and create right nowfor yourself?
Because I'm sure you do a lotfor other people, but if you
were to create something justfor fun that isn't out there
yet, what would you wannadevelop just for you?
Eric (31:27):
It's hard to say something that isn't out there unless it's
like magic, like Umer wassaying, teleportation and stuff
like that.
Yeah.
But what I really wanna build isso I feel for people, especially
new people they come to GPT.
It says Hey, ask me anything.
And they'll ask something anddon't get the result.
So I think it's very importantfor people to pay GPT.
Pay for GPT.
(31:48):
Yeah.
'cause of the models, right?
Like you get way better modelswhen you pay.
But people who haven't paid ormaybe they've paid, but they
come, they ask something, theydon't get the result they want,
and they're like, this shitdoesn't work.
Why would I pay?
Or whatever.
Brooke (32:00):
Yes.
So I really
Eric (32:01):
wanna build a platform that guides users through
prompts, using workflows, to getthe best output in images
instead of a cat sitting underthe tree, having some kind of
form.
They fill out like what style ofimage, what dimensions, what,
different stuff not only inimages but in text or video or
whatever, but just guideprompting users.
(32:24):
That's something I've beenwanting to build and there's a
lot of examples and I thinkthey're super cool.
So I just wanna do my version'cause I feel especially doing
all these workshops.
With The Collective, right?
I come into the event thinkingI'm gonna teach an automation or
something, and then I realizepeople can't even copy paste
prompts.
Brooke (32:40):
Yeah.
Eric (32:40):
Like they have a lot of trouble.
Brooke (32:42):
Yeah.
Eric (32:43):
Getting their desired output.
And even trying to teach themhow to prompt like the anatomy
of a prompt and stuff it's notonly hard for people, but
sometimes they don't even wantthat.
They don't wanna learn theanatomy of a prompt.
Just want their image, right?
Yeah.
Or they just want whatever.
So I wanna build a platformthat.
Guide prompts users to get theirbest result.
Brooke (33:02):
I've noticed that as well.
'cause I work a lot with clientsand teaching them the basics of
AI right now, and I've noticedthat some of them just want done
for you solutions.
But I do think it's stillimportant to start to learn it
in some basic elementary waybecause it's a skill that
everybody needs to know,especially just something as
simple as prompt engineering orjust how to speak to AI in
(33:23):
general.
I always recommend to say, okay,this is my end goal on the
prompt, and then.
To follow it up by saying, okay,ask me anything you need to know
to give me my end results that Imight have missed.
And that helps a little bit withthat solution of it, it being
like, oh, actually we also needthis, and this background and
(33:45):
context so that helps as well.
I also feel like that would bejust beneficial coming from a
developer because even when Igenerate images on ChatGPT, I
feel like I'm not doing a greatjob with describing and
prompting, but if I were in adeveloper's or a creative design
mind, I would know betterquestions to answer.
(34:06):
So that's a really cooleducational moment.
Do you like teaching?
Eric (34:11):
I think I love it.
I'm just getting started, right?
Yeah.
I'm just getting started.
I started doing speaking atevents this year.
I never thought I'd do it.
I've always wanted my YouTubechannel, but I never got started
'cause I'm camera shy.
But after these events I feellike can be a good teacher.
Before, I didn't know that.
I didn't feel that.
Brooke (34:29):
Yeah.
Eric (34:29):
But now, after receiving feedback and stuff I think I
could do a good job, especially'cause I learn all day, right?
All day, every day.
I'm on YouTube, so I know howthings, or I consider, I know
how things should be explainedto people who don't know ai.
Mm-hmm.
Or, Or even for devs, right?
I wanna teach.
Yeah.
A hundred percent.
I think.
That, guide prompting platformalongside a video vault.
(34:53):
I just wanna do a video vault,right?
'Cause there's two types ofvideos I would do.
One is what's new in AI everyweek.
'cause it never stops, right?
Never stops.
20, 30 different stuff comingout every week.
And a tool or different toolspeople can use instead of just
like using ChatGPT ask meanything, like having optimal
(35:14):
outputs.
I really wanna do that.
Brooke (35:18):
I would tune in, although I don't know if I'd be
able to keep up.
I've tried learning from you acouple times and it just, it was
a little difficult for N8N forme.
Eric (35:28):
I think it's also, it's difficult to try to teach
something in an hour.
Yeah.
To a group of 30, 40, 40 peoplewhere there's a lot of that
disparity gap is so big, right?
Somebody might catch it,somebody might not, but I think
in video format where people cango back a hundred times if they
need to, but if by the end ofthe video they'll get what's
(35:50):
promised, I think that's wayeasier.
That and one-on-ones.
'cause everybody learnsdifferent, right?
Yes.
So I think one-on-ones wouldmake it way easier than trying
to learn in group.
Brooke (36:02):
That's interesting to think about because as I train
teams, I do find that there is abit of disparity to your point,
and then sometimes one personjust has a question that goes
like completely off script andthat derails the training a bit.
So one-on-one actually soundslike a great use case.
(36:25):
So when you watch your YouTubevideos, you just pause when you
have a question or figure itout.
Eric (36:30):
Yeah.
I pause until I get that part ofthe step.
I also feel in groups, let's saywe're teaching automated
LinkedIn posting.
Some people don't even useLinkedIn.
So they'll try to follow alongjust so they feel they learn
something, but it's not the sameenergy they put into it as.
Where when I go on YouTube I'mlooking for what I wanna learn.
(36:52):
I'm super hooked on the video.
But having to come up withsomething that will interest
everyone.
Yeah.
Super hard.
Brooke (36:59):
Yeah.
Just today I was editing avideo.
I was trying to align audio andvideo onto script and I came to
a point where I needed like veryspecific help and I had ChatGPT
link, a YouTube resource, and Ieven said do not give me a 45
minute video.
Give me like short to the pointexactly the problem I'm having.
And it gave me two differentoptions.
(37:19):
It gave me an Instagram videoand a YouTube video, and it was
like, which one do you likeprefer?
It's been getting so smart inthat way.
Eric (37:26):
Super smart.
Brooke (37:27):
Yeah.
And I was able to watch thetutorial and solve the problem
within 20 minutes.
Eric (37:32):
That's crazy.
Yeah.
'cause another thing I don'tknow is people who use chat GPT.
I don't know how many of thoseusers.
Actually realize how betterthey're getting or if they just
take GPT for granted and yeah,feel it's the same GPT as day
one,'cause it's really not.
No, it's super crazy.
And open AI has to do a betterjob naming their models.
(37:53):
'cause right now, like everyone,so you go into ChatGPT on the
top left, you will always seethe model you're using and it
says GPT 4o.
Yes, it's their worst model.
But how would people know that?
Unless you click and then yousee o1 o3, but you say, ah, four
is bigger than three, right?
So it's probably better, butit's actually the worst.
So they gotta do a better job.
(38:14):
I agree.
In naming conventions, and also.
A better job explaining whateach model does.
Reasoning models versus nonreasoning models.
Doing a better job explaining.
'cause every time we explainthis to people and show them the
difference, they're like, oh,that's their oh crap movement.
Yes.
Brooke (38:31):
A very key point.
What do you feel is a keytakeaway that you wanna share
with anybody that's getting newto AI and just jumping in for
the first time?
That's a lot of my audience ispeople just starting to learn.
As a developer, do you have anyword of advice?
Eric (38:49):
Be open to learning.
Just be open to learning.
Yeah.
Yeah.
Keep in mind like this issomething that is already
changing the whole world.
Don't be scared.
Don't be scared and don't benegative about no.
It's taking everybody's job andI think you have to adapt.
And you gotta get started.
And you gotta have fun.
You're gonna have fun.
'cause once you start usingthese tools and see how they
(39:10):
benefit you personally.
Brooke (39:12):
Yeah.
Eric (39:12):
You won't hate it.
You'll love it.
Brooke (39:15):
I love that feedback.
You said that you are lookingfor more clients and thinking
about running ads this is youropportunity for a shameless
plug.
How can listeners reach out toyou?
Eric (39:29):
Just message me, I'd say on Twitter.
That's something I really wannago.
I'm not big into personal brandsor stuff, but Twitter Yes.
I'm saying Twitter.
'cause you need to get intoTwitter to stay up to date.
And I personally use Twitter allday, so I'll always see the
messages.
But yeah, seriously, YouTube andTwitter.
Getting your algorithms rightinstead of them showing you.
(39:51):
Unnecessary stuff.
Yeah.
Or just wasting your time.
Like you can really takeadvantage of that tech, which is
also ai to make your algorithmsbetter and just learn in small
videos.
Even a tweet, like reading atweet will make you learn crazy
amounts of stuff.
Brooke (40:05):
Wait, how do you use AI to make your algorithms better?
Eric (40:09):
So you, you start scrolling on Twitter, searching
for AI or whatever, let's say asan example, or crypto.
And then tweets you findinteresting.
There's different weights on ifyou like post or you retweet a
post.
Or you bookmark bookmarks arethe highest.
Yeah.
So if you start bookmarkingstuff, the algorithm will start
showing you more of that stuff.
(40:29):
Yes.
So bookmarks are the number one.
Brooke (40:33):
I knew that about Instagram, but I actually don't
have an X account
Eric (40:40):
or TikTok.
No comment.
Okay.
I don't have TikTok but x.
It's a little Gotta have it.
Gotta have it.
Brooke (40:46):
Yeah.
Gotta have it.
Do you use LinkedIn?
Eric (40:48):
No.
That's for boomers.
I mean the, the best, The bestjob opportunity I've had came
through Twitter.
Brooke (40:55):
Really?
Eric (40:55):
Yes.
Brooke (40:57):
How did that happen?
Someone just saw something thatyou shared volume, you for one
and your project
Eric (41:01):
launched.
And I found some bug on theirwebsite.
So I reached out and say Hey,this is not working.
This is not working.
I'm a dev.
I could fix it.
The guy was like, send me yourcv.
And I'm like, what is a cv?
I'll send you my portfoliowebsite, and I sent it and he
was like, okay.
And it's been my biggest andbest project.
Brooke (41:21):
Wow.
That's amazing.
One thing that we didn't touchon was crypto.
Actually, now that you justmentioned it, I don't know if
that's something that you wannachat about at all, if it's
relevant to AI in particular.
But I know you're verypassionate about that space, so
I just wanted to open the floorif anything that you want to
share about.
Eric (41:40):
The intersection of AI and crypto I feel is super
important.
'cause right now AI is superbeginner.
But as it gets more advanced andyou start actually hiring ai,
not only creating your customGPT, but having AI do their
tasks for you AI agents can'topen a bank account.
You can't pay them pesos ordollars or euros or whatever.
(42:02):
So the way, the only way we'regonna be able to transact with
ai, like paying them andreceiving payments from them is
gonna be.
Crypto and blockchain.
So I feel, that could be a wholeother episode Yes.
On crypto and Bitcoin butbasically I feel it's important
for people to maybe not fullyunderstand blockchain, but just
(42:22):
keep in mind, AI won't want yourdevaluating money.
They're gonna want crypto and itis gonna be the only possible
way to interact with them in amonetary way.
So buy Bitcoin.
Brooke (42:39):
Buy Bitcoin.
Yeah.
This is your plug.
Not
Eric (42:41):
financial advice, but buy Bitcoin.
Brooke (42:44):
What do you think about people who use AI to invest and
give them advice on that?
Do you think that's legit?
Eric (42:50):
I hope.
Yeah.
I hope'cause going back to yourquestion on what I would love to
build for myself would be acool, trading bot.
There's millions of trading botsout there.
I think eventually my predictionis eventually AI will become so
good at trading that everythingwill just collapse.
Like all the aas will be so goodthat I don't know how the
(43:11):
world's gonna work.
But if you don't know how toinvest, getting advice from AI
is probably good.
I don't know if I'd rely ahundred percent of it, but, me
personally, I would heavily relyon it.
'cause I'm not a savvy investor,right?
Yeah.
I'm just like buy Bitcoin hold,yeah.
But if you wanted to trade andstuff like that, probably
there's algorithms and stuffthat can do a better job.
Brooke (43:33):
I dunno if I would feel confident in it,
Eric (43:37):
but like you feel confident in AI creating images
for you, right?
Yes.
And drafting emails.
Yes.
And a lot of stuff like.
Why?
Why would trading be different?
Brooke (43:48):
I don't know.
I guess'cause it's more weightto it.
Eric (43:52):
You can start small.
Brooke (43:53):
Yeah.
Maybe you can start small.
That's how I approached cryptois just with small money.
Eric (43:58):
Yeah, small money.
And then if it's doing a greatjob, you can scale up.
I don't know if it's there yet'cause a lot of people would
just rely on ai.
I do think it's gonna get there.
No problem.
Brooke (44:08):
Interesting.
What is your Twitter handle sopeople can reach out to you?
Eric (44:13):
It's ErikfazJU E-R-I-K-F-A-Z-J-U.
We can link it at the bottom ofthe podcast.
Brooke (44:21):
Yes, we will.
Thank you so much, Erik.
I appreciate your time.
Thank you.
And it was an honor to be yourfirst podcast ever.
Eric (44:31):
Yes.
It's an honor for me.
Thank you a lot for theinvitation.
Brooke (44:35):
Yeah, you did such a great job.
Eric (44:36):
Thank you.
Thank you.
I appreciate it.
Wow, I hope today's episodeopened your mind to what's
possible with ai.
Do you have a cool use case onhow you're using AI and wanna
share it?
DM me.
I'd love to hear more andfeature you on my next podcast.
Until next time, here's toworking smarter, not harder.
See you on the next episode ofHow I Ai.
(44:58):
This episode was made possiblein partnership with the
Collective ai, a communitydesigned to help entrepreneurs,
creators, and professionalsseamlessly integrate AI into
their workflows.
One of the biggest game changersin my own AI journey was joining
this space.
It's where I learned, connectedand truly enhanced my
(45:19):
understanding of what's possiblewith ai.
And the best part, they offermultiple membership levels to
meet you where you are.
Whether you want to DIY, your AIlearning or work with a
personalized AI consultant foryour business, The Collective
has you covered.
Learn more and sign up using myexclusive link in the show
notes.
Welcome to How I AI the podcast featuring real people,
real stories, and real AI inaction.
I'm Brooke Gramer your host andguide on this journey into the
real world impact of artificialintelligence.
For over 15 years, I've workedin creative marketing, events,
and business strategy wearingall the hats.
(00:22):
I know the struggle of trying toscale and manage all things
without burning out, but here'sthe game changer, ai.
This isn't just a podcast, How IAI is a community.
A space where curious minds likeyou can come together, share
ideas, and I'll also be bringingyou exclusive discounts, free
trials and insider resources soyou can test drive the latest
(00:45):
tools and tech yourself.
Because AI isn't just a trend,it's a shift.
The sooner we embrace it, themore freedom, creativity, and
opportunities we'll unlock.
How I AI is brought to you inpartnership with The Collective
designed to accelerate yourlearning and AI adoption.
I joined the collective and it'scompletely catapulted my
(01:07):
learning, expanded my network,and show me what's possible with
ai.
Whether you're just starting outor looking to refine your AI
strategy, the collective givesyou the resources to grow.
Stay tuned to learn more at the end of this episode, or
check the show notes for myexclusive invite link..
Hello everyone.
Welcome to another episode ofHow I Ai.
(01:29):
I'm Brooke Gramer, your hostToday I have a very special
guest.
His name is Erik Faz Jurgensen,and he's a builder, he's a
visionary mind, and he's thefirst developer I've had on my
podcast, so I'm excited to divedeep into his journey and his
background.
Erik, welcome.
Thank you so much for beinghere.
Eric (01:49):
Thank you for having me, Brooke.
Brooke (01:51):
Yes, and I love to leave the floor open for you to share
more about yourself.
Eric (01:56):
Okay, so I'm a software developer.
I've been coding for the pastsix years almost.
Mm-hmm.
I have my own business calledSpy Dev, which is a one man.
I'm a serial freelancer.
Brooke (02:08):
Okay.
Eric (02:08):
I just have an a brand name behind it, but yeah I just
code all day.
I love coding.
I was a huge gamer before Istarted coding.
And then I saw YouTube video onhow to gamify your life and get
dopamine hits out of simplestuff the same way you do in
video games.
So coding I literally haven'ttouched a game controller in
(02:31):
five years.
Brooke (02:32):
That's so fun.
One YouTube video changed yourwhole trajectory.
Eric (02:37):
Yes.
Brooke (02:38):
My brother's a gamer.
He loves that space as well.
He's also very smart.
He's a background in engineeringand works in cybersecurity.
So share more about that journeyof what was like the first
initial step and what did youstudy in school exactly and how
did you really get into thisspace?
Eric (02:56):
So in school I was studying, innovation and
development engineering.
And I actually took some codingclasses and I hated it.
Oh.
I thought I was gonna likecoding.
I really, I hated coding and Iwould cheat on my exams to pass
coding classes, and then Istarted making websites.
And started running my solobusiness and eventually dropped
(03:17):
out from college.
And then a friend of mine whohas a big company in Mexico,
they raised money and he waslike, Hey, I want you to join my
team.
I want you to code with us.
And I'm like I don't code, Idon't know how to code.
And he's like, that's all right.
We'll teach you.
Wow.
So I accepted the gig and theirway of teaching me was just,
okay, these are the programminglanguages.
(03:40):
This is what you have to do.
Look it up.
So I started looking things up,like handing something in, and
then they were like, okay, nowdo this.
Now do this.
And I started having fun and Istarted coding that way.
And since then I've just workedon a lot of different crypto
projects.
(04:00):
Mainly crypto projects, codingin Web3.
Crypto changed my life forever.
The space, I didn't knowanything about it until I
started coding inside of it.
And yeah, been part of hugeprojects in crypto and now of
course AI is part of my, like Ido AI projects, but mostly I use
AI all the time.
Brooke (04:21):
How many projects do you have at any given time?
Eric (04:24):
So the most I've had in any given time is 23.
Brooke (04:27):
23,
Eric (04:28):
23 websites.
Wow.
Right now, I guess it's five orsix.
Okay.
Brooke (04:33):
Yeah.
Do you primarily work onwebsites or what kind of
projects do you feel like youspecialize in?
Eric (04:39):
Before it was websites.
Now I actually don't like makingwebsites that much.
I like to do like webapplications and more robust
systems.
The best project I'm working onright now, the one I enjoy the
most is called Veeam.
It's like my full-time job.
Okay.
And we're basically creating a,an AI platform where people can
generate video memes with ai.
(05:01):
So we're training some modelsand we're getting a lot of
context from clips, from memesand using that to create video
memes.
Yeah,
Brooke (05:11):
I love how.
You have fun with it.
And that's how you startedlearning was from a place of
just excitement.
Eric (05:18):
It's excitement.
And also I truly believe ingetting paid to learn.
'cause that's how I started.
So now whenever a new clientcomes in and says Hey, I wanna
build this platform, even if Ihave no idea how I'm gonna do
it.
I'll just say yes and learn.
So by the end of that project Ilearned a lot.
I got paid.
Mm-hmm.
I had fun.
So.
(05:39):
It's cool.
Brooke (05:40):
So do you ever take on projects where you don't know
how you're going to solve theirproblem or complete the project?
Eric (05:47):
All the time.
Brooke (05:47):
All the time.
All the
Eric (05:48):
time.
Brooke (05:49):
And how do you learn, how do you figure that out?
Eric (05:51):
YouTube.
YouTube.
YouTube.
Yeah.
Before it was YouTube I creditYouTube.
As my reason for dropping out.
Brooke (06:01):
Wow.
I call
Eric (06:01):
it YouTube University.
'cause literally YouTube haseverything, not only in coding,
whatever you wanna learn.
Yeah.
Changing the tire to coding.
It's all on YouTube and thenfirst it was YouTube, then GPT.
Brooke (06:15):
ChatGPT is how you learn everything now.
You just ask it what you need toknow.
Eric (06:19):
Now I don't learn.
Now it does the work for me.
You know?
Brooke (06:22):
It does the work for you.
Yeah.
Most of
Eric (06:23):
the work.
Yeah.
Brooke (06:24):
That goes to my next question is what are you using
right now?
What's your technology stack?
What are your favorite?
I.
Tools or do you create your ownin-house for everything?
Eric (06:34):
So my, my favorite tool, hands down is this app called
Cursor.
Cursor, which is what we call anIDE.
I don't know what it stands for,but basically it's like my
coding tool where I write codeand it was the first of its kind
that integrated ai.
So now I have a sidebar uh, withAI that has all of my project as
context.
So if I tell it, Hey, now createthis new tab, or this new page
(06:57):
for the project, or whatever, itknows that design I'm using the
colors fonts, styles, everythingI already have in the project.
So it'll do a great job.
So I mainly use cursor.
I pay for GPT for my sisters touse it.
Mm-hmm.
Uh, I cursor is my everyday app.
Everything app.
I just code all day and I justpay for different subscriptions
(07:18):
to try out 11 labs, Leonardodifferent apps, just to like a
new thing comes up, I'll pay forit, test it, forget I
subscribed, pay it a couple moremonths and then unsubscribe.
Brooke (07:30):
That was my next question is how much do you
think you're spending a month onai?
Eric (07:35):
So thankfully my projects pay for my stack.
Okay.
Most of my stack, yeah.
I just pay for like differentsubscriptions I wanna have fun
with.
But I'd say probably a hundredbucks a month.
Brooke (07:46):
Okay.
Eric (07:46):
Which is not that much.
Brooke (07:47):
No, not at all.
And it
Eric (07:48):
could actually be$20.
And I could do the same workwith 20.
'Cause cursor, what's greatabout Cursor is it has ai but
it's not specific to one model.
So inside of Cursor, I can useGPT Claude from Anthropic,
Gemini, from Google X, AI fromTwitter or X.
So I got all in one for 20 bucksa month.
(08:09):
Wow.
And.
It's crazy.
Brooke (08:12):
That's cool.
So what else are you excitedabout learning right now?
What is new to you in thisspace?
Eric (08:19):
What's exciting and new for me is catching up, right?
Catching up.
Catching up.
Like I, I feel I'm superadvanced, but at the same time,
being at the forefront ofeverything that's happening
gives me like anxiety.
'cause you never you don't knowwhat is better than what, right?
Last week, Claude four wasreleased, and then today GPT 03
Pro was released.
(08:39):
And they made 03 super cheap fordevelopers.
But then tomorrow, GoogleGemini's newest model work will
come out and they're all reallygood.
The hard part and the excitingpart at the same time is
figuring out which one best fitsmy needs in coding.
Brooke (08:58):
Do you feel like there's a lot of.
Overlap or they're still uniquein what it is that they do for
you?
Eric (09:04):
For me, I think there's a lot of overlap.
A lot of overlap.
I just go through Twitter andsee what other coders are saying
about the models and okay, nowI'll use Claude.
And if I start seeing, if myalgorithm starts saying oh no,
Gemini is now the best.
I'll just switch to Gemini.
But realistically all of them.
Do a great job.
And it's crazy, like workingwith them every single day
(09:27):
actually makes me notice howgood and how better they're
getting over time.
Literally by the day.
Yeah.
Like yesterday, there's a bug Icouldn't solve with Claude
today, it'll solve it, yeah.
They just keep getting betterall the time.
It's scary.
Brooke (09:41):
Do you have a community of coders and developers that
you're in?
Nah.
Do you feel like you just chatwith people on like Twitter?
I
Eric (09:48):
just, yeah.
Yeah, I just chat with people onTwitter.
Brooke (09:52):
Yeah.
Eric (09:52):
Yeah.
Because before there was thisthing called Stack Overflow
where whenever you had aquestion or something you
couldn't fix by yourself, youwould post a question post your
code, and a community of coderswould help you.
Now, ai.
It is my community.
I think I, I talk to the robots.
Brooke (10:10):
Yes.
We're in a community together,the collective, and we just had
a
Eric (10:14):
Mastermind.
I, we're not all coders, that'swhy I didn't say that.
We're not coders.
Yes, it's not a coder community,
Brooke (10:18):
but it's more of an AI in general community.
We recently had a mastermind andHacker House and you're
absolutely the most favoriteperson there because what does
everyone need to have theirdreams come true?
Developer, someone who knowscode.
So I'm curious because everybodyhas an app idea.
Everybody has a business idea.
(10:39):
Everybody has a super uniqueproblem that they're solving in
their personal lives, and thenext step is always how do I
make this come to life?
How do I.
Build.
And the next step is obviouslyto work with a developer or have
a CTO in your company.
How does someone go aboutprocuring that?
(11:00):
What would be the best way?
And speaking from a developer,how do you like to work with
people?
How do you approach thatrelationship?
Eric (11:08):
So I really like working freelance based.
Not getting a full-time job Yes.
Or something that will make meonly work in that project.
Brooke (11:15):
Okay.
Eric (11:15):
cause now I think it's a win-win.
'cause now we devs can workfaster.
Way faster.
You don't have to hire a team offive devs.
Mm-hmm.
You can have one or two devs.
And if you allow them to work inother projects we'll deliver
faster, right?
We'll do your project, we'll doit fast so we can move on to the
next project.
Brooke (11:32):
Okay.
Eric (11:33):
So I really like that.
And from the employer's point ofview I think it's super
important right now to focus onsearching for devs or people
that.
That are hungry to stay up todate.
Mm-hmm.
Right.
So before you needed seniordevelopers mm-hmm.
with 15 years of experiencestuff like that.
Right now, it's funny'causethose same guys are the ones
(11:55):
with the most resistance to ai.
They're so pro, they have theirown way of doing things.
And they're super expensive.
And they all work in Facebook,apple, Netflix, Google.
Yes.
Now like junior developers orpeople who haven't even
graduated can get way more workdone with way better quality.
I believe, than senior devs.
Brooke (12:15):
Wow.
Eric (12:16):
So that shift is crazy.
And I'd say just look for peoplethat, that wanna stay up to
date.
'cause if you're staying up todate, the tools do the most part
of the work mm-hmm.
right now.
And it doesn't have to be superexpensive.
You can get your MVPs or ideasbrought to life super fast,
super, super fast.
Wow.
And allow the devs to work ondifferent projects.
Brooke (12:37):
Yeah.
That's awesome.
One thing that we chatted abouta bit last week was vibe coding.
If you could share in your ownwords what it is and what you
see the future of Vibe Codingand ai.
Eric (12:50):
Yeah, so Vibe coding is what we now do every day as
coders.
So instead of writing code,we'll type in natural language
like, Hey, please add threecards, or Please change the
background color to red, orwhatever.
AI will do the job.
So that's why it's called Vibecoding.
And then you just review.
You have to do minor changes,like super small changes.
(13:11):
'cause the models are supergood.
But life coding is basicallyinstructing ai on what you need
or what needs to get done.
Yeah.
And then just overseeing that.
And there's a ton of apps.
So inside of Cursor, what we dothere, there's a lot of browser
apps like Bolt.new Yes.
Lovable.dev.
Yeah.
v0.dev from Vercel.
(13:32):
Those are all great, but I thinkthose are great for beginners.
Because you just type your idea,it'll bring it to life.
But once you go into the codingTrenches Cursor does the same
thing.
Whatever you can do in thoseapps, you can do in Cursor, it
just looks a bit moreoverwhelming'cause you do need
to know more stuff.
But it's super cool'cause youcan change the models and.
Let's say I ask GPT to fixsomething and it gets it wrong,
(13:55):
so I can revert and now tellGemini the same prompt, until
one of them gets it right.
And then I'll just stick tothat.
Yeah, that's pipe coding.
Brooke (14:02):
I wanna touch a little bit more about the anxiety of
being into this space.
I thought that was like veryvulnerable of you to share just
'cause on from the outside wewould think that developers feel
very confident and they're veryexcited and just sharing how, oh
my gosh, there's so much to stayon top of, and feeling like, you
know, all the right things.
(14:24):
Do you see that going away, ordo you think it's gonna just
always be this rat race oftrying to stay on top of
everything?
Because
Eric (14:32):
I think it'll get harder actually.
Yeah.
So what's crazy is I was blessedto use Chat GPT the day it came
out.
Wow.
I was scrolling through Twitter,saw this new tech.
Started using it, not even forcoding, just asking stupid
questions.
Yeah.
And I was like, wow, what isthis?
And then we started using it forcoding.
And we were like, wow what isthis?
(14:53):
But when that came out I don'tremember the exact roles or the
exact order, but basically wethought developers and
designers, out of all the jobsin the world, were gonna be.
The ones taken last by ai.
And it turned out to be theother way around.
AI first came for designers.
You can create images, coolvideos, cool websites all
through prompts.
(15:13):
And then it's coming for usdevs.
Mm-hmm.
Because now like vibe coding,people who have never coded in
their lives can do apps.
Can do websites in minutesbefore it took weeks, maybe
months if it's an app.
And also two or three years ago,computer science was the most on
demand skill.
(15:34):
Yes.
Two or three days ago, news cameout that it's the number one
unemployed wow.
University degree right now inthe us, like from being the
number one employed to the leastemployed.
Wow.
It like computer science is nowworse than arts and history.
Wow.
(15:54):
Which is crazy.
Which is crazy.
That's part of the anxiety,right?
You've gotta either keep up or,though AI won't take your job,
but someone who uses AI will.
That's uh, saying I'll neverforget when I read it.
So it's anxiety, it's fun.
'cause if you stay up to date,it just keeps getting better.
But for people who refuse to usethe tools to try new things out
(16:15):
they're gonna get wiped.
Brooke (16:17):
For instance, someone who vibe codes but has a
computer science degree is gonnabe able to do it way better.
Yeah.
Than someone who doesn't.
So I still think it's importantto have this background and
skill and knowledge, whetherit's YouTube University or real
university.
Eric (16:31):
Yeah.
But probably for a year.
Brooke (16:33):
Just a year.
Eric (16:34):
I think maybe less, like right now, yeah.
You do need a dev.
Let's say you vibe codesomething in Bolt and you wanna
add your specific domain to it.
You probably, you can YouTube itor you can have a dev do it.
In the future, all of thesetools are getting way better.
Like before it would only do thedesign.
Now it does the code.
Now you can integrate Stripe forpayments and you can integrate a
(16:54):
database.
Before it was just the frontend.
So now you can deploy directlyto GitHub, which is the Google
drive for devs.
So every week all of these toolskeep getting better and I feel
that gap from having to have acomputer science degree and not
having it.
It's shrinking so fast.
Brooke (17:13):
So it's valid to not be going to school for that.
Eric (17:17):
Yeah.
Brooke (17:18):
Wow.
Yeah.
So what, I think
Eric (17:19):
it's valid not going to school for anything, to be
honest.
Brooke (17:23):
Yeah.
What do you think, what do youthink people should be learning?
What do we need to know?
Eric (17:26):
I think people should learn how to learn.
Brooke (17:28):
Learn how to
Eric (17:29):
learn how to learn.
Whatever it is you wanna learn.
Yeah.
Just have the passion and thediscipline of Google it, you
know?
Mm-hmm.
or now Chaskt it that was SamAltman's way of Google it for
ChatGPT Mm-hmm.
Chaskt it
Brooke (17:42):
Chaskt I haven't heard that one.
Eric (17:44):
Yeah.
Chaskt it YouTube, it, Googleit.
All the infos out there.
And i, don't like universitiesbecause there's no way they
would be teaching everythingwe're learning right now.
Yeah.
For example, everything wetalked about in the hacker
house.
Yeah.
Not only us, like everything youcan find on YouTube, I feel
universities have their boomerway of teaching.
And super methodological, likestep by step and using AI is
(18:08):
cheating.
And they make you study on yourown.
Do homework on your own, likeyou can collaborate'cause that's
cheating.
Tests are on your own.
But then you go to the realworld and it's just teamwork
everywhere.
So I need to encourage everyoneto drop out.
Actually, I need to do it.
'Cause everything you wannalearn, you can learn it way
(18:29):
faster, way better outside ofschool, in my opinion.
Brooke (18:33):
Do you work in a team?
Eric (18:34):
I work in several teams.
Yeah.
Brooke (18:37):
That's awesome that you have that support and you made
such a valid point.
We only rarely do team projectsin school where it's like the
opposite.
It's mostly teams and then it's,
Eric (18:48):
you need a team.
You need a team in the realworld.
Like I, I am a solopreneur.
But I think, what's that sayingthat says you can go faster by
yourself mm-hmm.
but further with a team.
So I do think it's all aboutteamwork.
Funny enough, I think AI can bepart of that team, if not most
of the team.
(19:09):
Your
Brooke (19:09):
co-pilot.
Eric (19:10):
Your co-pilot.
Yeah.
And co-pilots.
Co-agents.
New Tech is coming out forbasically everything.
Um.
And it's just weird'cause I'mreading a book called Who Not
How, and it talks about theimportance of having teams.
Yes.
But this was written, I don'tknow how many years ago, and I
just keep asking myself okay,valid.
It's true.
You have to set bigger goalsthat you can't achieve by
(19:30):
yourself.
But how many of those who's cannow be different AICUs?
Because you can literally use itfor everything.
I use it for code, but you cando whatever.
You can run ads, you can dodesign, social media content.
Yeah, automate everything.
I don't know.
It's scary, but fun.
Brooke (19:45):
What does your AI copilot team look like?
Do you have a bunch of agentsworking for you at one time?
Eric (19:51):
Yeah.
Yeah.
So my team would be, I have mypersonal manager who's ai.
Okay.
He's the only agent that talksto me.
Okay.
I only talk to him and based onwhat I ask it or what I say that
needs to be done, it'll talk toa team of AI agents and delegate
those tasks to each agent.
'cause in the AI world, it'ssuper important to have,
(20:11):
specialized agents.
Not having one charge GPT thatsays, ask me anything.
And it'll do everything.
So if you train each agent to bevery good at emails, very good
at ads, at coding, whatever, andyou have one personal assistant
that is in charge of delegatingthe tasks to them that's my
team.
Brooke (20:29):
And how did you build out your team?
Eric (20:32):
With a lot of tools.
I love N eight N.
Okay.
We went over that in the hackerhouse.
Yes.
I forgot to mention Cursor.
N eight N is super great.
It allows you to not only buildagents, but create automations
with those agents.
So I've built a team with Neight N also an app called
Relevance ai.
Which is basically for that tobuild your AI team.
(20:52):
And actually the idea of.
Having a personal assistant andthen delegates tasks to other AI
team members came out from aYouTube tutorial.
Wow.
On relevance ai,
Brooke (21:03):
do you still have that YouTube tutorial?
Of course.
Maybe I'll link it in the shownotes.
Eric (21:07):
Yeah,
Brooke (21:07):
that's a really good one.
And so when you wanna talk toyour agent manager, how do you
talk to it?
Eric (21:12):
You can send emails, you can send it a WhatsApp message.
You can do telegram.
I like Telegram better.
'cause WhatsApp like Facebook orMeta has a lot of permission
issues.
And you have to have a businessand verify the business and get
approved and stuff.
Telegram, you can just connectit or you can even connect one
or several agents to Slack.
(21:32):
So you can send a Slack messageto your personal assistant and
then have it delegate the tasks.
So it can be the trigger, as wecall it can be anything.
You can imagine something inyour CRM an Instagram message,
whatever.
Okay.
Brooke (21:48):
Cool.
And so it just reports back toyou when the project's finished
or it needs additional support?
Eric (21:53):
Yeah.
You can either enable YOLO mode,which is you don't need any
approval from me, just get thetasks done and post or do
whatever you have to do.
Yeah.
Or you can have approval, right?
You can have approval by thepersonal assistant, or you can
also integrate to your team.
Agents that verify that the workis being done correctly, right?
So the agents will talk betweenthemselves make edits or
(22:17):
whatever needs to be done.
And then post, I still likeapproval.
I think it's important'cause younever know Yes.
When the agents can hallucinateand just start doing stuff on
their own, which can be reallydangerous.
You can approve, you can get anemail report or a message or
whatever that just says, this isdone, this is the post.
Do you like it?
Yeah.
Boom.
And it'll post.
Brooke (22:37):
I've heard that there's a workaround for hallucination
using the RAG system.
Do you use that?
Eric (22:43):
I do use it, but I I don't think it solves hallucination.
What RAG does is basically, it'sa better way of giving your
agents knowledge.
Brooke (22:52):
So
Eric (22:52):
They will hallucinate less on the knowledge they're trained
on, but that doesn't meanthey'll hallucinate less on the
actions they're taking.
For example, we built a reallycool agent and he had replied to
new emails that were coming.
Basically, it replaced perfectlyusing rag, but it started
replaying two emails from threeyears ago.
Brooke (23:10):
Oh no.
Eric (23:11):
And stuff like that.
So RAG helps for, from aknowledge standpoint, but not
like action.
Brooke (23:18):
Yes.
I think if I haven't made agentsyet, but I feel like I would
like to have the approvalprocess.
I wouldn't go YOLO mode.
Eric (23:24):
Yeah.
I don't do YOLO mode even incoding.
Okay.
I always have to approve evenwhat sources is the agent gonna
search?
Yeah.
Or what, and I mean, it's bittedious, having to click, accept
or approve or whatever.
Mm-hmm.
But it just makes it better.
You lose, you use less tokens.
It's less expensive.
Okay.
And it gets it right more thanif you just do YOLO mode right
(23:47):
now, I think in a few months youcan just have your mode on
everything.
Brooke (23:53):
Love dining out.
Here's a little gift for you.
I've been using InKind and it'shonestly a game changer for food
lovers.
It's like a universal diningwallet that earns you 20% cash
back at amazing restaurants,cafes, and bars nationwide.
And as I thank you for being alistener, you can grab$25 off
your next$50 tab when you signup with my link in the show
(24:16):
notes.
Just pay with InKind, earnrewards and use them on your
next outing.
A quick heads up.
This offer is for new InKindusers and covers food, drinks
and tax, but no tips or fees.
The credit expires 30 days afteryou claim it, so don't let it go
to waste.
Check the show notes for myaffiliate link and enjoy your
(24:38):
next meal on me.
Well kind Of.
So with everything that you knowin this space and its
adaptability and its ease ofuse, what do you think a lot of
people are sleeping on right nowwhen it comes to the benefits of
ai?
Because even I feel like.
I'm only tapping into 2% of whatI could be doing.
(25:00):
What do you think is somethingthat people could be doing
beyond ChatGPT as a beginner,what would you recommend they
could be immediately using forai?
Eric (25:11):
I think what people sleep on the most is, realizing they
can use AI for literallyeverything.
Brooke (25:16):
Yeah,
Eric (25:16):
literally everything.
What do you wanna do?
For example, for the podcast,you wanna run ads for the
podcast, you wanna reach out topeople.
Like anything you can imaginecan be done with ai.
Mm-hmm.
But I feel there's a lot ofresistance in my personal
experience.
I've seen a lot of resistance onpeople saying if I start using
AI for this and this, then I'mgonna stop using my brain.
And I'm gonna become obsolete.
I'm gonna become lazy.
(25:38):
Yeah.
That's one common denominatorI've heard in everywhere.
Like here in Bali or in the USor in Mexico.
They're like I don't wanna stopusing my brain.
Yeah.
And it is just super funny'causeI don't feel like it's, you're
not stopping your brain, you'reactually automating repetitive
tasks and liberating your brainto think about new stuff.
(25:59):
Yeah I just question everything,right?
Question everything, everythingyou're doing, whether you like
it or you don't okay.
How could AI do it or how couldAI help me do it better?
I think it's a mind shift that'shard to get to, but I think it's
necessary'cause tech's moving sofast, right?
Yes.
And I feel like the tech that'sout there right now is just the
(26:21):
tip of the iceberg.
And they have new tech ready,but they haven't released it.
'cause they first want people tostart using GPT for most things.
And start learning how the worldis gonna work before they just
automate your life, wow.
So I think stop havingresistance and question
everything.
Question everything.
And for people who have thisresistance I was talking to a
(26:42):
friend and my friend was sayinghow ai, artificial intelligence,
what's artificial is not theintelligence artificial are the
tools we're using to talk tothis intelligence.
'cause this intelligence hasbeen out there.
He compares it to Mother Nature,right?
There's intelligence everywhere.
In plants.
In trees, yes.
Like everywhere.
And we're just creating thetools to tap into that
(27:06):
collective intelligence.
So I think it's inevitable andpeople should, instead of being
scared learn to have fun withit.
Brooke (27:13):
That's really good advice.
Do you feel like that's how youapproached it in the very
beginning?
A
Eric (27:18):
hundred percent.
Yeah.
I've been so excited ever sincesince day one, super excited.
And I, and even I I use it forcoding, but I feel I should use
it for way more stuff likerunning ads.
I've never run ads in my life.
Never.
And I need ads to grow mybusiness or whatever I want do.
Yeah.
I should start using it evenmore.
Brooke (27:37):
You can clip up this podcast and run some ADs
Eric (27:40):
I should.
Brooke (27:41):
I should.
I should.
Eric (27:43):
'Cause I feel like once I do it and I start getting
results, it'll be like, wow.
'Cause three years ago since Iwanted to run ads, I would never
do it.
'cause I had to learn ads andlearn AB testing and learn have
someone that creates the copy,have someone that generates the
image.
Someone that's a master like admanager.
Now a, I know ai.
(28:04):
Can't do this, and I'm saying itso I hold myself accountable.
That I need to run.
At least thats, there's a lot ofstuff I could use AI for Yes.
That I should use AI for, but atleast running ads or for
example, I hate being in frontof cameras.
And I really wanna start myYouTube channel.
And right now it's so easy toclone yourself.
And create all of thesetutorials, right?
I don't even have to be in thecamera and I'm still putting it
(28:26):
off and I know it's easy.
I've cloned myself.
I just gotta, I don't even haveto write the scripts.
I just gotta get started.
So another piece of advice to meand to everyone is just get
started.
Whatever your idea is, yourproject.
You just gotta get started.
And I think you're a really goodexample.
I've been talking about thispodcast with the guys and it's
(28:47):
crazy how you just said, youknow what, I'm gonna start a
podcast.
And now I don't know whichnumber of guests I am, but you
got started, right?
Yeah.
You got started.
And that's super important.
'cause now I feel creativity isdemocratized.
Like now.
Now there's no.
Doing whatever is not hard.
(29:08):
You just gotta get started.
I just gotta get started.
Brooke (29:10):
Just get started.
Yes.
This past week together with allof us coming together for the
conference we were in, reallyinspired me to do more with the
podcast and be doing morepromotions and getting out there
more.
So I'm excited to just beputting more energy into it and
I don't know, I think you'relike, guest 15 by now.
(29:31):
That's, yeah.
It's cool.
That's fish.
Eric (29:34):
Super.
No, congrats.
That's crazy.
Yeah.
Brooke (29:36):
Thank you so much.
So you spoke a lot about thebenefits.
I'm curious, have there been anydownfalls or Oh crap moments
with using ai?
I think it's fun to hear likehow we've had learning moments
with using ai.
Eric (29:51):
Yeah, I feel coding projects has become so easy
using AI that we almost take itfor granted.
Yeah.
And then when we actually pushthe code to production, to live
mode.
We don't realize, like we don'teven double check now.
We're so confident.
It's doing a great job.
Sometimes we don't even check.
And let's say it fixed a hugeproblem.
(30:12):
So now that problem's fixed, wepushed that, but we didn't
realize that fixing thatproblem, it messed five things
up.
Oh, or the videos are notworking or whatever.
Oh.
And it has happened more thanonce.
Especially in a team.
'cause it's not only me, it'sthree of us.
Let's say we are all using ai.
I don't know what these otherguys are coding, for example.
(30:33):
So I might use AI to fixsomething missing up their work
without ever realizing,'cause Inever touched their code.
So yeah, I think that's whyhaving YOLO mode is a huge
learning moment.
But I think that's the only, oh,crap moment.
I fucked things up for the code.
You can always revert, right?
But yes, sometimes it's not usthat realize.
(30:54):
It's clients sending messages,like horrible user experience in
this part, and we're like, oh,shit.
That wasn't a problem yesterday.
That, that's the only,
Brooke (31:02):
yeah.
So nothing too bad.
My next fun question for you is,since you're in that space of
creating so much as a developer,what are you wanting to wave a
magic wand and create right nowfor yourself?
Because I'm sure you do a lotfor other people, but if you
were to create something justfor fun that isn't out there
yet, what would you wannadevelop just for you?
Eric (31:27):
It's hard to say something that isn't out there unless it's
like magic, like Umer wassaying, teleportation and stuff
like that.
Yeah.
But what I really wanna build isso I feel for people, especially
new people they come to GPT.
It says Hey, ask me anything.
And they'll ask something anddon't get the result.
So I think it's very importantfor people to pay GPT.
Pay for GPT.
(31:48):
Yeah.
'cause of the models, right?
Like you get way better modelswhen you pay.
But people who haven't paid ormaybe they've paid, but they
come, they ask something, theydon't get the result they want,
and they're like, this shitdoesn't work.
Why would I pay?
Or whatever.
Brooke (32:00):
Yes.
So I really
Eric (32:01):
wanna build a platform that guides users through
prompts, using workflows, to getthe best output in images
instead of a cat sitting underthe tree, having some kind of
form.
They fill out like what style ofimage, what dimensions, what,
different stuff not only inimages but in text or video or
whatever, but just guideprompting users.
(32:24):
That's something I've beenwanting to build and there's a
lot of examples and I thinkthey're super cool.
So I just wanna do my version'cause I feel especially doing
all these workshops.
With The Collective, right?
I come into the event thinkingI'm gonna teach an automation or
something, and then I realizepeople can't even copy paste
prompts.
Brooke (32:40):
Yeah.
Eric (32:40):
Like they have a lot of trouble.
Brooke (32:42):
Yeah.
Eric (32:43):
Getting their desired output.
And even trying to teach themhow to prompt like the anatomy
of a prompt and stuff it's notonly hard for people, but
sometimes they don't even wantthat.
They don't wanna learn theanatomy of a prompt.
Just want their image, right?
Yeah.
Or they just want whatever.
So I wanna build a platformthat.
Guide prompts users to get theirbest result.
Brooke (33:02):
I've noticed that as well.
'cause I work a lot with clientsand teaching them the basics of
AI right now, and I've noticedthat some of them just want done
for you solutions.
But I do think it's stillimportant to start to learn it
in some basic elementary waybecause it's a skill that
everybody needs to know,especially just something as
simple as prompt engineering orjust how to speak to AI in
(33:23):
general.
I always recommend to say, okay,this is my end goal on the
prompt, and then.
To follow it up by saying, okay,ask me anything you need to know
to give me my end results that Imight have missed.
And that helps a little bit withthat solution of it, it being
like, oh, actually we also needthis, and this background and
(33:45):
context so that helps as well.
I also feel like that would bejust beneficial coming from a
developer because even when Igenerate images on ChatGPT, I
feel like I'm not doing a greatjob with describing and
prompting, but if I were in adeveloper's or a creative design
mind, I would know betterquestions to answer.
(34:06):
So that's a really cooleducational moment.
Do you like teaching?
Eric (34:11):
I think I love it.
I'm just getting started, right?
Yeah.
I'm just getting started.
I started doing speaking atevents this year.
I never thought I'd do it.
I've always wanted my YouTubechannel, but I never got started
'cause I'm camera shy.
But after these events I feellike can be a good teacher.
Before, I didn't know that.
I didn't feel that.
Brooke (34:29):
Yeah.
Eric (34:29):
But now, after receiving feedback and stuff I think I
could do a good job, especially'cause I learn all day, right?
All day, every day.
I'm on YouTube, so I know howthings, or I consider, I know
how things should be explainedto people who don't know ai.
Mm-hmm.
Or, Or even for devs, right?
I wanna teach.
Yeah.
A hundred percent.
I think.
That, guide prompting platformalongside a video vault.
(34:53):
I just wanna do a video vault,right?
'Cause there's two types ofvideos I would do.
One is what's new in AI everyweek.
'cause it never stops, right?
Never stops.
20, 30 different stuff comingout every week.
And a tool or different toolspeople can use instead of just
like using ChatGPT ask meanything, like having optimal
(35:14):
outputs.
I really wanna do that.
Brooke (35:18):
I would tune in, although I don't know if I'd be
able to keep up.
I've tried learning from you acouple times and it just, it was
a little difficult for N8N forme.
Eric (35:28):
I think it's also, it's difficult to try to teach
something in an hour.
Yeah.
To a group of 30, 40, 40 peoplewhere there's a lot of that
disparity gap is so big, right?
Somebody might catch it,somebody might not, but I think
in video format where people cango back a hundred times if they
need to, but if by the end ofthe video they'll get what's
(35:50):
promised, I think that's wayeasier.
That and one-on-ones.
'cause everybody learnsdifferent, right?
Yes.
So I think one-on-ones wouldmake it way easier than trying
to learn in group.
Brooke (36:02):
That's interesting to think about because as I train
teams, I do find that there is abit of disparity to your point,
and then sometimes one personjust has a question that goes
like completely off script andthat derails the training a bit.
So one-on-one actually soundslike a great use case.
(36:25):
So when you watch your YouTubevideos, you just pause when you
have a question or figure itout.
Eric (36:30):
Yeah.
I pause until I get that part ofthe step.
I also feel in groups, let's saywe're teaching automated
LinkedIn posting.
Some people don't even useLinkedIn.
So they'll try to follow alongjust so they feel they learn
something, but it's not the sameenergy they put into it as.
Where when I go on YouTube I'mlooking for what I wanna learn.
(36:52):
I'm super hooked on the video.
But having to come up withsomething that will interest
everyone.
Yeah.
Super hard.
Brooke (36:59):
Yeah.
Just today I was editing avideo.
I was trying to align audio andvideo onto script and I came to
a point where I needed like veryspecific help and I had ChatGPT
link, a YouTube resource, and Ieven said do not give me a 45
minute video.
Give me like short to the pointexactly the problem I'm having.
And it gave me two differentoptions.
(37:19):
It gave me an Instagram videoand a YouTube video, and it was
like, which one do you likeprefer?
It's been getting so smart inthat way.
Eric (37:26):
Super smart.
Brooke (37:27):
Yeah.
And I was able to watch thetutorial and solve the problem
within 20 minutes.
Eric (37:32):
That's crazy.
Yeah.
'cause another thing I don'tknow is people who use chat GPT.
I don't know how many of thoseusers.
Actually realize how betterthey're getting or if they just
take GPT for granted and yeah,feel it's the same GPT as day
one,'cause it's really not.
No, it's super crazy.
And open AI has to do a betterjob naming their models.
(37:53):
'cause right now, like everyone,so you go into ChatGPT on the
top left, you will always seethe model you're using and it
says GPT 4o.
Yes, it's their worst model.
But how would people know that?
Unless you click and then yousee o1 o3, but you say, ah, four
is bigger than three, right?
So it's probably better, butit's actually the worst.
So they gotta do a better job.
(38:14):
I agree.
In naming conventions, and also.
A better job explaining whateach model does.
Reasoning models versus nonreasoning models.
Doing a better job explaining.
'cause every time we explainthis to people and show them the
difference, they're like, oh,that's their oh crap movement.
Yes.
Brooke (38:31):
A very key point.
What do you feel is a keytakeaway that you wanna share
with anybody that's getting newto AI and just jumping in for
the first time?
That's a lot of my audience ispeople just starting to learn.
As a developer, do you have anyword of advice?
Eric (38:49):
Be open to learning.
Just be open to learning.
Yeah.
Yeah.
Keep in mind like this issomething that is already
changing the whole world.
Don't be scared.
Don't be scared and don't benegative about no.
It's taking everybody's job andI think you have to adapt.
And you gotta get started.
And you gotta have fun.
You're gonna have fun.
'cause once you start usingthese tools and see how they
(39:10):
benefit you personally.
Brooke (39:12):
Yeah.
Eric (39:12):
You won't hate it.
You'll love it.
Brooke (39:15):
I love that feedback.
You said that you are lookingfor more clients and thinking
about running ads this is youropportunity for a shameless
plug.
How can listeners reach out toyou?
Eric (39:29):
Just message me, I'd say on Twitter.
That's something I really wannago.
I'm not big into personal brandsor stuff, but Twitter Yes.
I'm saying Twitter.
'cause you need to get intoTwitter to stay up to date.
And I personally use Twitter allday, so I'll always see the
messages.
But yeah, seriously, YouTube andTwitter.
Getting your algorithms rightinstead of them showing you.
(39:51):
Unnecessary stuff.
Yeah.
Or just wasting your time.
Like you can really takeadvantage of that tech, which is
also ai to make your algorithmsbetter and just learn in small
videos.
Even a tweet, like reading atweet will make you learn crazy
amounts of stuff.
Brooke (40:05):
Wait, how do you use AI to make your algorithms better?
Eric (40:09):
So you, you start scrolling on Twitter, searching
for AI or whatever, let's say asan example, or crypto.
And then tweets you findinteresting.
There's different weights on ifyou like post or you retweet a
post.
Or you bookmark bookmarks arethe highest.
Yeah.
So if you start bookmarkingstuff, the algorithm will start
showing you more of that stuff.
(40:29):
Yes.
So bookmarks are the number one.
Brooke (40:33):
I knew that about Instagram, but I actually don't
have an X account
Eric (40:40):
or TikTok.
No comment.
Okay.
I don't have TikTok but x.
It's a little Gotta have it.
Gotta have it.
Brooke (40:46):
Yeah.
Gotta have it.
Do you use LinkedIn?
Eric (40:48):
No.
That's for boomers.
I mean the, the best, The bestjob opportunity I've had came
through Twitter.
Brooke (40:55):
Really?
Eric (40:55):
Yes.
Brooke (40:57):
How did that happen?
Someone just saw something thatyou shared volume, you for one
and your project
Eric (41:01):
launched.
And I found some bug on theirwebsite.
So I reached out and say Hey,this is not working.
This is not working.
I'm a dev.
I could fix it.
The guy was like, send me yourcv.
And I'm like, what is a cv?
I'll send you my portfoliowebsite, and I sent it and he
was like, okay.
And it's been my biggest andbest project.
Brooke (41:21):
Wow.
That's amazing.
One thing that we didn't touchon was crypto.
Actually, now that you justmentioned it, I don't know if
that's something that you wannachat about at all, if it's
relevant to AI in particular.
But I know you're verypassionate about that space, so
I just wanted to open the floorif anything that you want to
share about.
Eric (41:40):
The intersection of AI and crypto I feel is super
important.
'cause right now AI is superbeginner.
But as it gets more advanced andyou start actually hiring ai,
not only creating your customGPT, but having AI do their
tasks for you AI agents can'topen a bank account.
You can't pay them pesos ordollars or euros or whatever.
(42:02):
So the way, the only way we'regonna be able to transact with
ai, like paying them andreceiving payments from them is
gonna be.
Crypto and blockchain.
So I feel, that could be a wholeother episode Yes.
On crypto and Bitcoin butbasically I feel it's important
for people to maybe not fullyunderstand blockchain, but just
(42:22):
keep in mind, AI won't want yourdevaluating money.
They're gonna want crypto and itis gonna be the only possible
way to interact with them in amonetary way.
So buy Bitcoin.
Brooke (42:39):
Buy Bitcoin.
Yeah.
This is your plug.
Not
Eric (42:41):
financial advice, but buy Bitcoin.
Brooke (42:44):
What do you think about people who use AI to invest and
give them advice on that?
Do you think that's legit?
Eric (42:50):
I hope.
Yeah.
I hope'cause going back to yourquestion on what I would love to
build for myself would be acool, trading bot.
There's millions of trading botsout there.
I think eventually my predictionis eventually AI will become so
good at trading that everythingwill just collapse.
Like all the aas will be so goodthat I don't know how the
(43:11):
world's gonna work.
But if you don't know how toinvest, getting advice from AI
is probably good.
I don't know if I'd rely ahundred percent of it, but, me
personally, I would heavily relyon it.
'cause I'm not a savvy investor,right?
Yeah.
I'm just like buy Bitcoin hold,yeah.
But if you wanted to trade andstuff like that, probably
there's algorithms and stuffthat can do a better job.
Brooke (43:33):
I dunno if I would feel confident in it,
Eric (43:37):
but like you feel confident in AI creating images
for you, right?
Yes.
And drafting emails.
Yes.
And a lot of stuff like.
Why?
Why would trading be different?
Brooke (43:48):
I don't know.
I guess'cause it's more weightto it.
Eric (43:52):
You can start small.
Brooke (43:53):
Yeah.
Maybe you can start small.
That's how I approached cryptois just with small money.
Eric (43:58):
Yeah, small money.
And then if it's doing a greatjob, you can scale up.
I don't know if it's there yet'cause a lot of people would
just rely on ai.
I do think it's gonna get there.
No problem.
Brooke (44:08):
Interesting.
What is your Twitter handle sopeople can reach out to you?
Eric (44:13):
It's ErikfazJU E-R-I-K-F-A-Z-J-U.
We can link it at the bottom ofthe podcast.
Brooke (44:21):
Yes, we will.
Thank you so much, Erik.
I appreciate your time.
Thank you.
And it was an honor to be yourfirst podcast ever.
Eric (44:31):
Yes.
It's an honor for me.
Thank you a lot for theinvitation.
Brooke (44:35):
Yeah, you did such a great job.
Eric (44:36):
Thank you.
Thank you.
I appreciate it.
Wow, I hope today's episodeopened your mind to what's
possible with ai.
Do you have a cool use case onhow you're using AI and wanna
share it?
DM me.
I'd love to hear more andfeature you on my next podcast.
Until next time, here's toworking smarter, not harder.
See you on the next episode ofHow I Ai.
(44:58):
This episode was made possiblein partnership with the
Collective ai, a communitydesigned to help entrepreneurs,
creators, and professionalsseamlessly integrate AI into
their workflows.
One of the biggest game changersin my own AI journey was joining
this space.
It's where I learned, connectedand truly enhanced my
(45:19):
understanding of what's possiblewith ai.
And the best part, they offermultiple membership levels to
meet you where you are.
Whether you want to DIY, your AIlearning or work with a
personalized AI consultant foryour business, The Collective
has you covered.
Learn more and sign up using myexclusive link in the show
notes.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.