#33. Synthetic AI agents

Episode Summary

In this episode I speak with Ahmed Reza about AI and synthetic agents.

Episode Notes

Ahmed’s background, Darpa, DoD, Airlines, startups

First company helping Dentists, what Ahmed learned and how he transferred that to Yobi

What Yobi does today, digital twins, synth agents, etc

Why coding will no longer be important, with Natural Language AI

Future of communication with AI

How this will affect early stage investments, what startups will find valuable

Why life is so special and unique


Text Ahmed Synth Agent: 6672207187

Frank Sinatra Sings “Get Low”:

Episode Transcription

Welcome everyone to another episode of Prompt Pod, an open-ended 
exploration into the field of AI and my quest to document conversations with 
bright minds in the space. I'm your host, Danny Kirk, and today I'm joined by 
Ahmed Reza.
Ahmed is the c e o and founder of Yobi app, a rapidly growing AI 
communications app for business with a background in AI and software 
engineering. Ahmed started his career developing image processing software for
NASA's Spitzer Space Telescope at Cornell University. He also participated in 
the DARPA Grand Challenge, contributing to the development of the world's 
first self-driving cars after [00:01:00] successful exits from various software 
Ahmed co-founded Trip Hub. A nonprofit open source version of Y Combinator
supporting over 30 startups currently based in Silicon Valley. Ahmed is de 
dedicated to leading Yogi app and driving innovation in an industry with ai. 
Ahmed, welcome to the show. Thanks so much for having me. On this very 
windy day, I'm glad we were able to do it.
So, uh, let's go back some years here. Could you tell me, um, kind of before 
your, some of your fir first successful startups, tell me about DARPA Grand 
Challenge, d o d airline technology, and also did you have any inklings of 
entrepreneurship early on in your life? Yeah, I think, uh, I used to hide it 
because I didn't want people to think that I wasn't smart.
Uh, because if you were an engineer and you were always hustling on the side, 
you know, people wondered. Uh, I started out a business called Ahmed House 
of Essence back at the Kga Mall when I was [00:02:00] going to school, uh, in 
Ithaca at Cornell. And, uh, it's kind of stuck with me always. There's, there was 
always a side hustle, but once those two things came together, it was amazing.
and p a part of me thinks that that's actually the reason that's Barrett, that grit, 
that just, you know, insatiable, uh, thirst for adventure. Is what got me into 
things like the DAR Grand challenge. Um, I was walking across the engineering
quad when I saw this, uh, little race car from the S SAE team, and I asked them 
if I joined the team, could I drive the car?


They looked at me and they said, not with how fat you are. So after that, uh, I 
looked for what other teams were there? Right down, right down. Uh, a few 
boots was. The self-driving cars. And I was like, well, I'm a geek, and that 
sounds really cool. So I really wish I had the foresight to know I was working 
on these self-driving cars in the future and, uh, the team there [00:03:00] made 
no, uh, they made no, you know, a secret of the fact that there was a lot to be 
figured out.
You know, we, we were, our first milestone was getting the car outta the garage.
Incredible. And was that your first, um, exposure to AI working on DOPA 
Grand Challenge, or did, was it in some of your classes before that? So ai, like 
AI has evolved a lot. There's, there's, there's a lot to it. For me, I'm one of those 
people that learns by doing, so that was one of the first, uh, times we're thinking
about computer vision, looking at lidar, looking at how do you process all of it, 
and, uh, ai, the way that we know it today, where this thing that talks to you, it 
can do all these other things, right.
There are many different versions of it. So early on, the more rudimentary 
forms of ai, um, but very effective forms of AI have existed in your car's 
transmission. For example, since 1996, American cars [00:04:00] have had, uh, 
transmissions that learn over time. So that's one of the reasons my cars, uh, I 
drive differently after a while.
So the artificial intelligence part has been there, uh, since long before, you 
know, it's, uh, once it's no longer fascinating, it just becomes tech. And what 
kind of fascinated about you at first? I, I assume there was something that kind 
of kept you going along those lines. Y you know, honestly, I'm like, I'm a geeks 
geek, so.
When I worked, uh, on the Spitzer Space Telescope, the fact that I got to take a 
detector and go to a particle accelerator, which was like right there, right? And 
bombarded with alpha, alpha particles like that is just so cool, man. That is just 
so cool. It's hard to pass off something like that. So, uh, it wasn't necessarily just
But it was this technology evolving to me. Like I'm [00:05:00] still fascinated 
by computers and microchips and f PGAs, uh, which not a lot of, uh, not a lot of
high level programmers know about. But having that breadth of knowledge has 
really helped me in my career in, uh, even in my entrepreneurial career, being 
able to zoom, zoom out and look at the bigger problems that you're trying to 


And we've got a lot to cover here, but please mention briefly what you did with 
airlines, both on the QR code side of things and also wifi tech, because I think 
it's a couple things that all of us use that we have you to be thankful for. So it's 
funny, like I didn't know much about startups because I'm an immigrant and I'm 
an engineer, right?
I didn't know startups really existed. I joined a startup, I joined several startups, 
uh, and my second startup that I joined like must be the luckiest guy on earth to 
be joining startups that do well. I didn't know about it, but the second startup 
that I joined, uh, I remember during the interview. The [00:06:00] guy, you 
know, Chris Harney, he was interviewing me, he had a hole in his t-shirt and it 
was between that and uh, and a finance firm where they were very mean during 
the interview and they offered two and a half times a salary.
I went with a startup, I was like, I wanna hang out with the guys with a torn t-
shirt. And they're doing really cool things. I have fully stocked fridge and like 
dual screen monitors and that was a really good choice cuz that company was 
called Kinetics, which was later bought by npi. Uh, sorry, which was later 
bought by, uh, national Cash Registers.
Um, and when you go in to the airport and, you know, you put your credit card 
in the little machine that gives you prints out your boarding pass, that's what 
Kinetics built and they dominated the market. Uh, 80% of the market was ours. 
This was before getting acquired, and we wrote services in c plus plus or H P U 
Uh, You know, it was, it was insane. Like we built scalable systems before all 
[00:07:00] these books came out, and we did it all in c plus plus. I remember 
writing web services in c plus plus and, uh, just a couple of H P X servers were 
handling 80,000, uh, transactions per hour. Uh, and what they do is these, these 
servers sit in the middle of these ancient GDS systems and it's just an interface 
talking to these older systems and giving you a much nicer ui.
So if you've, uh, gone from, you know, typing things in, To going to the kiosk, 
to your mobile boarding pass where you get the 2D barcode that you scan. 
Right. Um, so I remember, I remember programming that the html emitter that 
would produce the barcodes and, and it's really cool. Every time I use it, you 
know, I get this little joy like, Hey, I, I built that.
Incredible. And then you were also working on right after that, uh, working with
JetBlue on their wifi in air wifi. Yeah, so the in-a wifi project was super cool


cause I'd got into the travel industry. [00:08:00] And, uh, I knew, I know an 
inordinate amount of travel because I had to learn all these codes and things.
When I joined JetBlue, I saw a whole different angle. Uh, well, it was actually a
startup, that jet rub robot, right? Um, this startup, uh, was called Live tv. They 
built these antennas, uh, on top of your aircraft that would get you television, 
live television, while you're up in the air. Uh, and then they retrofitted it with, 
uh, uh, with, uh, you know, modified antennas.
And then we launched new satellites, uh, so that we could give really high speed
internet in the air. And you had to overcome these crazy things that you don't 
have to overcome in the ground, like, uh, you know, your ping response time in 
750 milliseconds. So just because of the fact that you're going to a satellite from
the air to the ground.
So we had to l figure out very out of the box solutions, like faking the ax, faking
the ax back to your device. [00:09:00] So when you're open the air and you're 
getting 70 megabites per second internet, you're welcome. The reason it's not 
flooding the network it's slowing down is because you get fake acts back from 
the server that's on board.
You preload content and uh, it's just. It's just the kind of stuff that really, really 
like fascinates me and, uh, makes me excited to work on hard problems. In early
2012, you founded Dental Web. Now kind of your first, uh, biggest success as 
far as companies you've, uh, founded. Could you tell us a little bit about what it 
did, what you learned from it, and then what you then took, uh, to kind of, I 
guess, from those learnings to create Yogi app?
Uh, certainly. So dental web. Now the founding story of it is really fun cuz here 
I am, you know, I've been in part of these startups. I wanna, I wanna launch 
something on my own. I'm in Florida, there isn't much of a venture capital 
community around there. So I try to work my inner Steve Jobs. That's the 
I try to work my inner Steve Jobs. I'd go, Hey, what would Steve Jobs do? He 
would go talk to customers. So I talked to one of my friends who was a dentist. 
And I'm trying to build him like the coolest, smartest practice management 
system that there is when I see a $3,000 check on his desk for the yellow pages.
And I just felt really offended by this. Uh, I was like, dude, what's wrong with 
you? You're paying yellow pages $3,000 a year. He goes, no, I pay them $3,000
a month. And it was, that was it. I was like, why would you do that? And he


said, this is where my customers come from. And I just wanted to prove to him 
how, how horribly wrong he was.
So I built a call tracking system, right? Using Asterisk, I rolled a call tracking 
system that would intercept inbound calls and then would record them and 
forward them onto his front desk. And I did that for all of his, uh, marketing. He
was also doing Google AdWord and some SEO and some other stuff. And after,
[00:11:00] uh, after 90 days, I went back to him.
I had all these calls recorded, I had all this data, and I showed him that out of 
the thousand phone calls that he got, only 56 were from, uh, the Yellow Pages. 
And the rest of them were, uh, from Google AdWords and other places. And he 
had only, he'd paid $10,000 for five customers. He looked at that data, he turned
around and said, wonderful, you're gonna be my marketer.
To which I was really offended, but then he was like, you get the $6,000 a 
month. So I worked my inner Steve Jobs again, and that was the beginning of 
dental web now, and just through word of mouth referrals that grew really fast. 
So in retrospect, it was basically using machine learning to classify phone calls 
and figure out which phone calls drive you revenue and figure out customer 
acquisition costs, which is really like fundamental and basic.
And that grew, that grew way bigger than I ever thought. Uh, something could 
grow a bootstrapped, uh, I sold that in 2018 and realized if that [00:12:00] little 
bit of machine learning can help your average dentist make an extra 400,000 a 
year. Imagine how many other business functions there are. Imagine if instead 
of just the marketing calls, I could know all of their communication.
So I set out to build a brain in the cloud, uh, a brain in the cloud that would do 
that work for you. Um, which is what we were really doing, is we were doing 
the work of an analyst and a marketing optimization person, right? And we also 
started giving advice on how to answer the phone because, you know, 
answering the phone like, hello, please hold, Is just universally a bad idea, 
Especially when you spend 50 bucks or a hundred bucks to try to even get that 
phone call to, to get that phone to ring. So we made little popups that would pop
up in front of the receptionist to let them know that this person who's calling 
could potentially be a $50,000 customer. So like smile and be nice, right?
We even gave them scripts. So took all those learnings and realized that. We 
[00:13:00] could start there and create something super powerful. So we started


working on a brain in the cloud that would help businesses improve their 
marketing, sales, and customer service. And lo and behold, we have Ybi app, 
which brings everything together in one place.
And it's AI infused from the very get-go. From the very beginning. The 
intention was to use artificial intelligence to augment, to augment business, 
right? But I also made it with my friends in mind, like the kind of people who 
are really hardworking, who are very innovative, but they don't have time to go 
learn HubSpot.
They don't have time to go learn what a C R M is, even. But they understand 
like what as a really simple UI looks like, right? They, they can just go and see 
all of their, uh, all of their communications in one place. Your WhatsApp, your 
Facebook, your text messages. And your phone calls. Phone calls, which then 
get transcribed, analyzed, [00:14:00] and.
Then we went one step further with some of the more newer innovations in AI 
where we realized we could actually create these synthetic agents to handle a lot
of the work for you. Not just optimization, but even step in and do the work so 
that you have a much easier time scaling. So what do synthetic agents mean to 
Yogi app these days?
Specifically? What do they do and how are companies using them? So synthetic
agents are a very interesting concept, right? It's when you spool up an AI that 
acts like a human in some ways, right? And so I have a digital phone that 
sounds very much like me. I could probably do a much better job on the podcast
than I do, but I, so I have one synthetic agent that does pitches for me.
So if you text me on that number, uh, if you're a venture capitalist or you just 
wanna know more about yo, instead of me repeating myself a [00:15:00] whole 
bunch of times. It's able to go through, because a lot of business con 
conversations are pretty shallow, right? There are only so many reasons why 
you call a dentist.
There are only so many reasons why you call a plumber, right? Uh, you might 
build a relationship with Joe the plumber, right? Uh, but you're calling the 
plumber to get plumbing done. So in this case, Our, uh, synthetic agents are 
actually really well fit. So you see, um, you could technically call Chad Chippy 
t a synthetic agent, right?
Uh, if you gave it a name and you called it Bob, you could talk to Bob or 
Clippy. I don't know if you remember Microsoft's clippy, right? I do not. A very


smart, not a very smart one, but we, we call it synthetic Asian instead of bots 
because bots have this very robotic feel to it. But the whole proposition or the 
whole faith of AI has been, you know, with the generalized AI has been, if it 
can, if it can fool you, or rather it can convince you that you're talking to a real 
[00:16:00] person, like that would be a significant step, and that's what you can 
do today.
Is when you're talking to your dentist or your, when you're talking to me, uh, 
and you're reaching out, I, I have every intention of responding to you, but what 
I know is that I'm not gonna be able to respond to a hundred thousand people, 
but my synthetic clone can. And a hundred, a hundred thousand of those people 
generally have very, uh, very shallow questions.
Shallow meaning what's yogi about? Tell me more about X, Y, Z, tell me more 
about your thoughts on this. Uh, and we built a goal-oriented conversation 
engine. And the goal-oriented conversation engine is different from something 
like Chat G P T, which has really open-ended conversations. It's a lot more like 
a sales development rep whose job is to stay on point and to.
You know, close the deal. So if you talk to my synthetic agent, it will tell you 
about yo V. It'll try to get you an appointment with me. And, uh, if you ask it 
about, [00:17:00] you know, geopolitical structures, uh, it, it will probably 
respectfully decline and say, Hey, let's get back to talking about yo. Incredible. 
So what do you see the future of communication in business, uh, being with 
synthetic agents and things like that?
Does just this just get bigger and scale, or what's your viewpoint on that? So this
is a really interesting, you know, sorry to get super philosophical, but I can't 
help it, right? With, with this kind of tech, you just can't help it. But my dream 
To be able to focus on conversations like this. And that dream is very close to 
being realized where I'm not in a distracted state, where the synthetic Asian 
handles, you know, and kind of, uh, triages requests that come in and there's a 
lot that it can handle. There's a lot that my digital phone can handle that I don't 
have to.
And the few things that I do have to handle, it's [00:18:00] given to me in a 
much more organized way. Um, now if you think about it, people used to build 
houses with their bare hands, then they got some tools. Push was cool, but you 
still needed a lot of labor, right? Uh, those pyramids, they are, they're marvels 
because you look at 'em and go, man, that, that sucked for a lot of people.


So in incomes, construction equipment, the industrial revolution, right? What 
did we do with construction equipment? Did we say, Hey, everybody, like, you 
know, no work for you. You guys starved to death because you know, we no 
longer need you. No, we built skyscrapers. And that's, to me, the promise of AI 
is once you have synthetic agents, it's like supercharging the human being.
So instead of digging with my hands, I can dig with this machine, like where in 
one scoop it can do more than you could do in an entire day. So I, I, I see AI as 
accelerating business [00:19:00] forward giving, uh, giving those like really 
gritty entrepreneurs, this edge that they wish they had. Uh, it is going to be 
challenging for established players who've kind of built up their organizations 
Humans, right? Like one manager to seven reports, you know? And that kind of 
a structure, much of which is around human limitations. Now you can actually 
transcend those limitations and you'd have to reimagine processes in 
organizations. You had told me something pretty mind blowing that I had never 
really considered, but is now quite obvious that, um, code language is kind of a 
third party mutual language between humans and machines.
Neither of them naturally speak it. Uh, so that's the power of natural language 
software. Can you tell us a little bit more about your views? Well, about how 
that may mean that coding itself may no longer [00:20:00] be that important. 
Yeah. So if you, if you look at it, uh, having been a programmer for a long time,
uh, being a very proud member of the geek community, you know, uh, making 
fun of code or not co code that's not commented, you know, variable names.
So a lot of what we learn is actually not how to code so that the machine 
understands it. If you named all of your variables X, Y, X, Y, X, Y, Z, right? 
They would understand it just fine. But the reason you name your variables like 
first name, right? Underscore homepage, right? The reason you name your 
variables like that as so that the next programmer understands what the heck 
you were talking about, right?
So when you look at someone's code and you say, oh, that's really good code. 
Oftentimes it's because it's very legible, it's maintainable, all of these other 
things. Now, The computer doesn't really care. So one of the startups I worked 
at was [00:21:00] Super Chips. And at Super Chips we reversed engineered 
ECUs. Um, it's been long enough that I can say that now we hacked cars to 
make 'em go faster, and there were these guys that would stare at binary.


They would stare. It was like the matrix. We had these giant screens and these 
guys star at binary long enough. Uh, and using Ida Pro, they figured out which 
things were for loops, which things were what. So they reconstructed code and 
once in a while, even found bugs in general motors of code or Dodgers code and
would let them know that you have bugs over here.
And that's gotta be super creepy. But if you think about it, the, that's what the 
computer sees. The computer just sees code. And if you look at ai, how it's 
trained, it's not traditional programming, and this is why a lot of confusion 
arises is people go, we don't really know why it's doing what it's doing.
That's because it wasn't programmed, right? If you took a piece of metal and 
then you melted it [00:22:00] into a bowl, it's like you go, okay, that's, you 
know, here's the mold. This is why it's circular, right? Uh, and AI for, to, for 
some instances like taking a hammer and hitting that bull hard enough, enough 
times that it, uh, hitting that piece of metal hard enough that it turns into a bull, 
But you're not really sure exactly where you have to hit, right? It's not really so 
process oriented. So that's pretty cool. What's revolutionary about it is suddenly.
Now, uh, I have the output that I want. I know the input, and I know the output. 
And suddenly you can talk to this thing. You can talk to the machine.
If you can talk to the machine and it can just go right binary for you. It can go, 
what would you care? What would you care? How many four loops and how the
coding was like, that's for you, the human. It might explain to you that, hey, 
this, this could be the equivalent of three, four loops and you know, this other 
type of, uh, uh, pattern, [00:23:00] right?
But really there's a, there. Is a barrier right now between you and binary, right? 
And that's usually in c plus plus or Java or JavaScript, whatever, whatever level 
of abstraction you have. But it's not unimaginable that you could speak to your 
ai, your, your synthetic agent and say, can you check my emails every two 
And if I get an email from X, uh, you know, or can you check the internet for, 
uh, these five trends? And let me know when this happens. And then it talks 
back to you and say, why, uh oh, what's your motivation for doing that? And in 
a conversation, it ends up building this ex executable file that runs, that just 
does something perfectly right, really well optimized like that can actually 


So you don't necessarily need a programmer. Um, now I'm not saying 
programmers are going away like [00:24:00] tomorrow by any means, right? 
Uh, but I'm saying this is actually pretty revolutionary because from the very 
early days of programming to now from programming f BGAs and logic gates 
to seatbelt plus to, you know, very high level Python, right?
Um, I see this getting even further abstracted away to where you just speak to, 
to the computer. You know, it's kinda like the computer and Star Trek. You say 
computer do X and it just, does it, does that make it more successful or, um, 
accessible for most of the world, or what do you see the second and third order 
effects that be?
Absolutely. Absolutely. And that's what's really, that's what's really exciting is. 
At the very core of my being, I feel like I'm a problem solver. I love solving 
problems, right in cool new ways. Now I had to go learn a lot of things in order 
to solve those problems, and those resources wouldn't have been accessible to 
me had I not immigrated to the United States had I not gone to school where I 
went to school.[00:25:00] 
Um, but suddenly you democratize that. You don't necessarily even need to 
learn English. Which by the way is like my fourth language, right? That was a 
struggle just learning English, right? So imagine not having to know, uh, 
anything besides your own language. And now being able to speak to a 
computer that can teach you, that can code, that can build things, and you can 
then take.
Wisdom that you learned from wherever you are in the world and turn it and, 
and you can create solutions for the rest of the world. It's actually really 
fascinating to me. I can't wait for that future to happen. A lot of people, or not a 
lot of people, but um, some investors talk about how technology is, um, 
deflationary and I think a lot of people focus on just the technical aspect of it.
But honestly, when you make it so you can speak any language and you don't 
have to know coding, it's no longer the price of, uh, developer and. Living in 
San [00:26:00] Francisco, which is expensive to live in, that speaks English. It's 
um, anybody that speaks their own language that has access to um, AI anywhere
that can be doing those things.
And I think that's why it can be incredibly deflationary. Would you agree with 
that? Um, I think. The fundamental assumptions, so this is, this is gonna sound 
crazy, but the fundamental assu assumptions of a capitalist system or a 
communist system or whatever systems, I think this technology is so powerful.


We should ask ourselves those first principles questions is like, how do we 
organize society? You know, uh, our, our economics, everything that we live in,
the construct that we live in is very much shaped by the industrial revolution. So
here here's comes this next revolution in how humankind lives and is organized.
Um, yes, does expensive make sense in an AI [00:27:00] world, in a world of 
plenty? Uh, what, how do we organize ourselves? Do we get more towards Star 
Trek? And of course, my geek side is showing because I'd love to live in a Star 
Trek like Utopia. Uh, yeah, I guess what, what would you like to see is, would 
it, in your mind, would it be post scarcity economics?
Is that what we kind of end up going to, or what would you like to see happen? 
To be honest, um, I think we've been in this bo post scarcity world for a little 
while now. We're. We're kind of starting to catch up. Uh, we've been in an post 
scarcity world. When I was growing up, I actually saw people starve to death in 
It was one of the most horrible memories etched into my brain. I'll never forget 
it. Um, and people were lamenting that the population growth in the world was 
so big, so crazy that there's no way that we'd be able to feed this many people. 
[00:28:00] Look at us now. We're able to feed people. I mean, that is the 
promise of technology.
That's the promise of science. That's the promise of geeks. That's what we get 
up in the morning to do, is solve problems like feeding humanity. Check. That's 
awesome. Right? What can we do with this next level of technology? Because 
yes, we do create abundance, and the people who work on these things, people 
who work on vaccines, they do it because that's their legacy.
That's what they leave behind for humankind. Which is so much greater, so 
much more interesting than just having, you know, flashier spin or rims. You 
co-founded, uh, trip Hub, a nonprofit open source version of Y Combinator. 
Um, it seems like probably something of your legacy. What excites you about 
working with those startups?
I think it's just. Know, asking, getting to a point in my life where, uh, I wanted 
to think about what am I [00:29:00] doing, you know, what do I believe in? I 
talked to my mentor, Mike Milken, who was a billionaire, this after my dad 
passed away, and I said, like, Mike, I'm not nearly as wealthy as you are. I have 
a rolls Roy, and I feel really empty inside.


So he sits me down, looks me in the eyes, and he goes, the best thing you can do
with money, Is deploy it towards the way that you want to see, the, the way that 
you wanna see the world. You know, the changes that you want to see in the 
world. Like, what can you do? How can you be of service? You just need 
enough for a roof over your head and your basics.
And that, that level is pretty low. Beyond that, you go to ostentation, you're not 
gonna find fulfillment. it made me step back and go, what do I really believe in?
And I really believe in geeks. I really believe in geeks. I really believe in 
people, uh, like myself who are trying to problem solve, figure things out.
And having been a geek in larger organizations where oftentimes we're just 
looked at as tools, right? Like, uh, hey, go, go off and code this thing 
CodeMonkey, [00:30:00] right? And it, it's a really weird dynamic. Because a 
lot of geeks, at least I felt that my own potential was untapped. And uh, we 
initially started this thing coders, hackers, founders, and we called it Fight Club 
for hackers.
Cause we were worried we'd get fired if pe our bosses found out we were 
working on projects outside of work. Right. Getting together and. Um, we had 
no idea what the heck we were gonna build it. This was not for money. This was
just because we felt this need to build. Um, our first meetup was five people, uh,
and myself and the other two people were friends.
So it was really two people that showed up and that turned into a community of 
over a thousand in Melbourne, Florida very quickly. You know, coming 
together on nights, weekends, doing hackathons, building stuff. Before you 
know it, one of them got funded. Comes to us and we, we all pulled together our
money to actually rent out this first floor in downtown [00:31:00] Melbourne.
Uh, and one of the guys comes and comes in and says, Hey, I, I just got funded 
and it wouldn't have happened had it not been for this. So we were like, what do
we call it? So GitHub was, was really popular among geeks. We were like, well 
this is like, you know, entrepreneurship GitHub. So we called it Trap Short for 
Entrepreneur Hub and we built it as a nonprofit cuz.
We didn't really want to think about the profit side. We were just like, Hey, let's 
get this thing together. We were mostly on Hacker News. We really liked that 
spirit and that's how Trump Hub came about. And I'm very, very proud of the 
work, uh, that I've done at Trump Hub that others have done at Trumph Hub for 
the for-profits and the non-profits.


Like one of the nonprofits at Tripa by Keith Donald is called Steady Town. Um,
Keith Donald, uh, co-founded Spring Source before, so if you've ever used Java 
Spring, It was that dude, uh, he came into one of the meetups one day and he, 
you know, he, it was just this [00:32:00] other geek and he goes, Hey, I'd love to
buy you this building.
I'm like, what? Did you have money like that? And he's like, yeah, we did all 
right. You know, stole spring sweater. Then. Then I was like, wow. Um, Very 
humble guy. He still drives a Subaru. Sorry to put you on the spot, but, uh, you 
know, he, he comes in, he, he started this thing called Steady Town, which 
rehouse homeless people and he basically took his lessons from product 
development, really dove in.
Tried to figure out, analyze the problem of homelessness, subdivided it, and 
said, just gonna focus on the homelessness caused by economic, uh, you know, 
economic reasons, right? There are substance abuse and other, other issues. 
And, you know, it doesn't, I guess YC does nonprofits too now, but at the time, 
like it didn't matter.
What happened, but Steady Town has rehoused over a thousand families, which
is amazing. Uh, there was another one called Robotics in the Indian River 
Lagoon. Um, uh, the founder of that [00:33:00] group came to me and asked for
$5 million so that he could build a sub in, uh, in less than two years. Showed me
the plans and I was like, You're like totally overestimating my net worth, but 
I'm still building my own company at the time.
Like it's like, but let's put the, let's put the word out to our fellow geeks and see 
what can happen. I check in with them three months later. They're on the fourth 
iteration of the submarine. That autonomously cleans up gun out of the Indian 
River. And after their very first meetup, he said, some dude walked in with a 
submarine and said, I was working on this already, so here's a sub.
We can make this autonomous. And then they iterated on the design. And so it's
really crazy what you can do when you bring smart people together who want to
solve problems. They can't help it. Uh, they can't help but but solve problems. 
And that's kind of been my ethos. That's really what I believe in. And the more 
of that that I can do, the more of that [00:34:00] community that I can get to 
participate in, like the more fulfilling my life is.
Cause I believe that these are the kind of people that will change the world for 
the better. had mentioned that you're fascinated by life. Do you believe that 
we're creating a new li type of life form here with ai? And what fascinates you


about AI in general? I know a lot of people, uh, think that this is somehow 
immortalizing you, right?
Upload your brain to the cloud. No offense to the folks who do think that I am 
not one of those people. I believe life is ephemeral and that's what makes it 
really beautiful. The part about AI that, that's really fascinating to me is here's 
this machine looking back at you and it's, you have to reflect on what makes 
you you.
So the very first clone that I have is this better version of me as a salesperson. 
and you know, I can introduce you to my ai. So here you can talk to my 
synthetic agent. Hi, I'm Ed Yobe, a [00:35:00] synthetic agent, trained on his 
personal data and communications. It's nice to meet you all. I, so here's the thing
that sounds just like me.
Doesn't stumble, doesn't go, uh, doesn't curse, you know, is very perceptive and 
is better than me in so many ways. But at the end of the day, You're still 
probably going to like me and you're still probably going to love me because of 
the defects. And if you look at your relationships, what it means to be human, 
you love your friends because you know, they have this quirk, you know, you 
love them despite, their oddity.
And it's actually really. Amazing to watch people in their journeys as they 
become better, as they grow, uh, as they grow as people. So the drama of our 
life, I think, is absolutely ephemeral and beautiful. And it's beautiful because it's
ephemeral, it's beautiful, because it's temporary and, uh, I think [00:36:00] you 
can immortalize certain parts of yourself.
So if you were, um, if you were Aristotle, you wrote your thoughts down and it, 
you immortalized yourself in a sense. Today, the medium has just changed. You
can speak with the AI version of me maybe long after I'm dead, but you're not 
actually speaking with me. I'm, I'm this very defective. Bag of meat and water 
and whatever, right?
But this is me. You know, this is the me that my family loves. This is the me 
that, you know, people who know me know, uh, the imperfect version of me. 
What that is to me is like theorization, or that is just a new form of media. 
Where you can interact. So you can now talk to Aristotle and ask him questions 
instead of reading his book or instead of reading his tablet or instead of 
watching a movie.


Right. So like when movies came out, I mean, [00:37:00] must have blown 
people's minds like, oh my god, this, that's Aristotle. No, that's not actually 
Aristotle. Right. So now you can talk to Arisal and you can make him rap. Uh, I
heard Frank Sinatra doing some really horrible, like it was just, Unholy. Frank 
Sinatra was, uh, you know, a ad-libbing to, uh, some rap music.
That's much after his time. Is that Frank Sinatra? No, it's not right. It's just a 
different type of medium. So if you really wanted to know Frank Sinatra, you'd 
probably have to talk to his friends. And that is special. They're, they have this 
privilege that I will never have. Uh, and that's okay, but I can get to know him 
through his music, through his works, and through people making rap remixes 
with Frank and Ed.
Yeah, amazing. Very good points there. Always our final question. What 
interaction with AI or interaction that you've heard about with ai, have you 
laughed the hardest at. Oh my God. It was, [00:38:00] uh, it was hearing, uh, 
Frank Sinatra that was today actually was hearing Frank Sinatra say from the 
window to the wall, the,
so. I am, I am looking that up right now. I can send you the screen. I was like, 
I'm not sure you can undo it. Right. We, yeah, we will be linking to that in the 
show notes. Actually, I, I'd love to share that this has probably been the most 
fun that I've had, uh, in my professional career. Because once you start cloning 
voices, once you start putting two plus two together, you know, I'll go into 
Slack and there'll be like a message from me giving everybody the day off.
Or, uh, there'll be a message from Luke saying, uh, hey, thanks Ahmed. You 
know, I got your memo about being able to use the corporate credit card on 
Ferrari. These are just some of the smaller ones. Uh, I think [00:39:00] Hodge, 
uh, who's, who's on the product team? Created a picture of me, took one of my 
pictures, modified it to look like a Pharaoh, and put a caption below that said 
when the pyramids were due yesterday.
So it's like we get trolled and there's just, there's just so much joy in like, uh, in 
this new form of creative outlet, right? Uh, and you can generate these things so
fast. So I think it can, it can bring, it can genuinely make our ephemeral lives a 
lot more joyous, a lot more fun, a lot more creative. Um, if we just look at it, 
you know, if we take a slightly different cognitive frame, you know, I'd look at 
the world a little bit differently, maybe from the future.
Welled, thanks very much for coming on our show. If listeners are interested in 
learning more about what you're doing, where can I point them to online? Uh,


so I'm gonna do something different. Uh, of course you can find me on 
LinkedIn, you can find me on Twitter, and you'll realize soon that you're 
probably interacting with some part of a [00:40:00] synthetic version of me.
Uh, if you really want to talk to my synthetic agent, I'll give you the number. 
You can text 6 67. 2 2 0 7 1 8 7. Again, that's 6 6 7 2 2 0 7 1 8 7. If you text that
number, I will get back to you instantly, but understand, this is my synthetic 
agent so I can see what my synthetic agent is saying to you.
Please don't be abusive. That synthetic agent represents me. Amazing. I just 
wrote that down and we will be linking to that in the show notes. well, Annette, 
this was a pleasure. Thanks again for coming on the show. Pleasures all mine. 
Thanks so much and thanks to you, my dear prompter for tuning in, and I hope 
you enjoyed this conversation as much as I have.
If you enjoyed the show, please consider subscribing and leaving a good review.
Take care and always be prompting.