Fintech Without the Jargon: Making Healthcare Lending Accessible with AI w/ Sharmeen Aqeel, Founder & CEO @ Lyyvora

WEBVTT
00:00:00.000 --> 00:00:02.600
Initially, I estimated that it will take them
00:00:02.600 --> 00:00:06.459
300 hours to create the first portal. And when
00:00:06.459 --> 00:00:08.919
I showed them the portal, they gave me an estimate
00:00:08.919 --> 00:00:12.779
of 60 hours instead of 300 hours. So AI, where
00:00:12.779 --> 00:00:15.960
it can be a huge help to build product to go
00:00:15.960 --> 00:00:19.179
faster, it can also destroy your identity if
00:00:19.179 --> 00:00:22.780
they find you spam AI. So using AI, you need
00:00:22.780 --> 00:00:25.539
to understand how you're using it. Are you using
00:00:25.539 --> 00:00:27.719
it to build your image or destroy your image?
00:00:27.949 --> 00:00:30.769
One thing that I don't see AI is doing is the
00:00:30.769 --> 00:00:33.710
network that I'm creating. It's the people that
00:00:33.710 --> 00:00:36.390
I'm connecting today and who knows me, who know
00:00:36.390 --> 00:00:39.829
what I do. Welcome to Artificial Insights, where
00:00:39.829 --> 00:00:41.609
we have conversations with leaders and thinkers
00:00:41.609 --> 00:00:44.609
in AI about how AI is changing economics, work,
00:00:44.750 --> 00:00:48.850
and us. I'm your host, Daniel Maneri, and this
00:00:48.850 --> 00:00:52.049
is Season 5. Today, I'm joined by Sharmeen Akhil,
00:00:52.210 --> 00:00:55.689
founder and CEO of Livora. I was interested in
00:00:55.689 --> 00:00:58.210
talking to Sharmeen because she's doing something
00:00:58.210 --> 00:01:01.250
that is quite difficult to do with the current
00:01:01.250 --> 00:01:04.510
funding landscape. When you go and you apply
00:01:04.510 --> 00:01:07.329
for funding as a business, it seems like it's
00:01:07.329 --> 00:01:09.269
a lot of work to get all of your ducks in a row
00:01:09.269 --> 00:01:11.409
and present the right documents. And you might
00:01:11.409 --> 00:01:14.150
get rejected the first, second, or even a few
00:01:14.150 --> 00:01:17.189
times. And that often stops people from pursuing
00:01:17.189 --> 00:01:20.109
lending and getting a loan for their business
00:01:20.109 --> 00:01:22.290
that could help them grow, even if they could
00:01:22.290 --> 00:01:23.930
have just found the right lender and kept their
00:01:23.930 --> 00:01:26.030
business going. Charmaine's solving that problem.
00:01:26.510 --> 00:01:28.069
When I talked to her, I found it interesting
00:01:28.069 --> 00:01:32.290
that she said, yes, there's AI, but the big missing
00:01:32.290 --> 00:01:34.769
piece was technology to keep people together.
00:01:34.930 --> 00:01:37.689
In particular, that aspect of keeping people
00:01:37.689 --> 00:01:40.030
together, of bringing them into a community has
00:01:40.030 --> 00:01:43.510
side effects that even if ChatGPT were able to
00:01:43.510 --> 00:01:46.150
give you amazing advice on lenders as easily
00:01:46.150 --> 00:01:48.950
as it gives you advice on a sofa, you wouldn't
00:01:48.950 --> 00:01:51.310
have a community of people. This community that
00:01:51.310 --> 00:01:53.430
Charmaine is able to build because she is solving
00:01:53.430 --> 00:01:55.680
a problem in a specific domain. is something
00:01:55.680 --> 00:01:58.120
that could stand the test of time. Sharmeen,
00:01:58.219 --> 00:02:00.060
could you tell us a little bit about who you
00:02:00.060 --> 00:02:03.000
are and what you do? So Daniel, I am Sharmeen
00:02:03.000 --> 00:02:06.480
Akhil. I had been professionally product designer
00:02:06.480 --> 00:02:09.560
and working as a product design leader for last
00:02:09.560 --> 00:02:13.199
seven years. Recently, I started a company, Livora.
00:02:13.419 --> 00:02:15.759
It's a lending platform for healthcare clinics.
00:02:16.099 --> 00:02:19.259
And yeah, mother of two kids living in Calgary,
00:02:19.379 --> 00:02:23.560
Canada. Very cool. And could you tell us a little
00:02:23.560 --> 00:02:27.419
bit about... How that journey resulted in entrepreneurship,
00:02:27.840 --> 00:02:31.139
product design, starting your own company. That
00:02:31.139 --> 00:02:33.860
sounds like a pretty big jump. Yeah, it is. But,
00:02:33.919 --> 00:02:36.960
you know, it's also my history which made me
00:02:36.960 --> 00:02:40.180
do this bigger step. When I was a teenager, I
00:02:40.180 --> 00:02:43.620
wanted to become a decorator, like making things
00:02:43.620 --> 00:02:47.139
beautiful. And then I started, then I saw my
00:02:47.139 --> 00:02:49.990
school, Equal Design. I was in France. And I
00:02:49.990 --> 00:02:52.990
saw that they're taking design to solve human
00:02:52.990 --> 00:02:56.530
problems. At that time, it was in 2007, 2008,
00:02:56.870 --> 00:02:59.370
the human -centered design was not as famous
00:02:59.370 --> 00:03:03.110
as it is now, especially in France. I was so
00:03:03.110 --> 00:03:05.590
attracted that how they're using design as a
00:03:05.590 --> 00:03:10.129
tool to combine engineering, design, like statistic,
00:03:10.310 --> 00:03:13.870
and commercialization. Because all these three
00:03:13.870 --> 00:03:16.090
things together make something to make anything
00:03:16.090 --> 00:03:18.330
successful. So that was the basic thing that
00:03:18.330 --> 00:03:21.379
I was so attracted. But after six, seven years,
00:03:21.620 --> 00:03:24.979
I got into the school. I did my graduation. And
00:03:24.979 --> 00:03:27.759
for the sixth year, my school was teaching only
00:03:27.759 --> 00:03:31.280
to how to solve a problem. And not all the other
00:03:31.280 --> 00:03:34.180
things were optional. After that, I came into
00:03:34.180 --> 00:03:37.280
the user experience design, which is more common,
00:03:37.379 --> 00:03:40.039
which is more famous at that time. And then became
00:03:40.039 --> 00:03:42.939
a product designer. But I was always wanted to
00:03:42.939 --> 00:03:45.960
solve a problem through my design skills. So
00:03:45.960 --> 00:03:48.360
I had been doing it. But when I came in the leadership,
00:03:48.990 --> 00:03:51.889
Things started slowing down. I solved the problem
00:03:51.889 --> 00:03:55.250
for the team, but those problems were not big
00:03:55.250 --> 00:03:57.490
enough for me. And so this is where I started
00:03:57.490 --> 00:03:59.849
thinking that what am I doing with my life? I
00:03:59.849 --> 00:04:02.490
want to solve bigger problems. And then this
00:04:02.490 --> 00:04:05.389
is when I started finding this gap between financial
00:04:05.389 --> 00:04:09.370
world and the healthcare world especially, where
00:04:09.370 --> 00:04:12.689
there is an absence of technology. We can make
00:04:12.689 --> 00:04:16.970
things easier, but... There's so much of noise
00:04:16.970 --> 00:04:21.509
of jargon of financial things. And so I thought
00:04:21.509 --> 00:04:24.069
that maybe it's time for me, people like me,
00:04:24.129 --> 00:04:26.589
who can solve those problems through human -centered
00:04:26.589 --> 00:04:29.410
design. And so I started this company, Libora.
00:04:29.589 --> 00:04:33.329
And my first objective was, how can I make the
00:04:33.329 --> 00:04:37.389
loan application system easier for the borrowers?
00:04:37.449 --> 00:04:40.629
And so I created this app. I started my website
00:04:40.629 --> 00:04:43.209
form and then started this partnership and so
00:04:43.209 --> 00:04:46.579
on. And so today, yes, a founder hustle, it's
00:04:46.579 --> 00:04:49.579
different. It's overwhelming sometimes. But I
00:04:49.579 --> 00:04:51.639
know that today I'm working to solve a bigger
00:04:51.639 --> 00:04:54.920
problem. And there's a question that I think
00:04:54.920 --> 00:04:57.000
is on everyone's minds with a podcast like this,
00:04:57.100 --> 00:05:01.439
which is, Sharmeen, are you an AI? I am the AI,
00:05:01.560 --> 00:05:04.720
I would say. You know, I think this is another
00:05:04.720 --> 00:05:07.439
blessing that I got. When I was working in a
00:05:07.439 --> 00:05:11.339
company, AI was there in early 2025 and in 2024.
00:05:12.269 --> 00:05:15.449
The company was not letting me use the full potential
00:05:15.449 --> 00:05:18.990
of AI because we were allowed only JetGPT. And
00:05:18.990 --> 00:05:21.589
for anything else, we had to have this, the agreement
00:05:21.589 --> 00:05:24.470
of the company if I want to do. But since I started
00:05:24.470 --> 00:05:28.149
my own company, it's my rules. So I am just checking
00:05:28.149 --> 00:05:31.430
out AI from all different angles and just creating
00:05:31.430 --> 00:05:35.449
workflows. And I just can't imagine that I can
00:05:35.449 --> 00:05:38.370
do whatever I'm doing today without AI. Is that
00:05:38.370 --> 00:05:42.300
from a... entrepreneurial angle or is that from
00:05:42.300 --> 00:05:45.959
the business angle or both or more? It's both.
00:05:46.139 --> 00:05:49.959
So first of all, most of the entrepreneurs, they
00:05:49.959 --> 00:05:52.920
don't need a technical founder. They need someone
00:05:52.920 --> 00:05:56.560
who can help them. So for me, AI become my co
00:05:56.560 --> 00:05:59.720
-founder instead of co -pilot because for every
00:05:59.720 --> 00:06:03.920
different problem, I created one chat bot or
00:06:03.920 --> 00:06:09.500
chat GPT or another process, another agent. to
00:06:09.500 --> 00:06:13.060
solve my problem. And so it's my right hand now,
00:06:13.160 --> 00:06:16.139
AI. And in terms of business, yes, it's also
00:06:16.139 --> 00:06:19.800
giving you one example. I wanted to develop the
00:06:19.800 --> 00:06:22.379
prototypes for my software and being a designer,
00:06:22.519 --> 00:06:24.779
we are very handy. We go Figma and we started
00:06:24.779 --> 00:06:29.220
doing it. It took me seven hours to do my first
00:06:29.220 --> 00:06:32.879
workflow for the borrower application. But I
00:06:32.879 --> 00:06:34.540
was thinking that maybe there might be a better
00:06:34.540 --> 00:06:38.300
way. And finally, I found a software. which in
00:06:38.300 --> 00:06:41.660
70 seconds made a better job what I did in seven
00:06:41.660 --> 00:06:44.819
hours. It was like, oh my God. You know, it's
00:06:44.819 --> 00:06:47.199
because there's so many things which can be automated
00:06:47.199 --> 00:06:50.100
by AI and we don't need to do that. And we are
00:06:50.100 --> 00:06:52.279
wasting our time, literally wasting our time
00:06:52.279 --> 00:06:55.040
doing it by ourselves. Seven hours doesn't sound
00:06:55.040 --> 00:06:57.720
too long to me, especially compared to maybe
00:06:57.720 --> 00:07:00.040
what it would take someone with less experience
00:07:00.040 --> 00:07:03.279
designing things. Why did that feel long to you?
00:07:03.339 --> 00:07:05.569
What's the urgency? When you're a founder, you
00:07:05.569 --> 00:07:08.550
have so many things to do at the same time. When
00:07:08.550 --> 00:07:11.009
I was a designer, I was doing this design. So
00:07:11.009 --> 00:07:13.410
that was good for me. It's not much. But now
00:07:13.410 --> 00:07:16.290
I have to do marketing. I have to do the product
00:07:16.290 --> 00:07:19.949
development. I need to meet my students who are
00:07:19.949 --> 00:07:22.410
working with me to explain them how to do that.
00:07:22.769 --> 00:07:26.230
I'm also learning other tools like Excel. I'm
00:07:26.230 --> 00:07:29.089
making this debt plus MCA tracker easier for
00:07:29.089 --> 00:07:31.569
my users. And I'm understanding how to make.
00:07:31.920 --> 00:07:34.740
Excel, easy to use for the user. So there are
00:07:34.740 --> 00:07:37.600
hundreds of tasks going on during the day. Being
00:07:37.600 --> 00:07:40.360
a founder, I cannot afford to put seven hours
00:07:40.360 --> 00:07:43.220
in one task, especially when it can be automated.
00:07:43.839 --> 00:07:47.819
I like that urgency. And you mentioned that the
00:07:47.819 --> 00:07:50.839
business also benefits from AI. How have you
00:07:50.839 --> 00:07:54.019
used AI in the business or in the product itself?
00:07:54.439 --> 00:07:57.779
So the number one thing that we are doing is
00:07:57.779 --> 00:08:01.860
matching. the borrowers with the lenders where
00:08:01.860 --> 00:08:04.680
we are using ai in this particular area and i
00:08:04.680 --> 00:08:06.699
think that this is the best area where we can
00:08:06.699 --> 00:08:09.839
utilize ai is we collected the information of
00:08:09.839 --> 00:08:12.620
each lenders their credit boxes that what kind
00:08:12.620 --> 00:08:15.939
of borrowers they are interested in once we got
00:08:15.939 --> 00:08:20.540
those credit boxes we started creating the borrower
00:08:20.540 --> 00:08:23.759
profiles and also just to test how it's working
00:08:23.759 --> 00:08:27.889
and through ai we have generated how We will
00:08:27.889 --> 00:08:30.069
be able to match borrow with multiple lenders
00:08:30.069 --> 00:08:33.009
and give the borrower the best option. So this
00:08:33.009 --> 00:08:36.009
is number one where we are using AI for our business.
00:08:36.269 --> 00:08:38.669
But there is one more thing that I'm using in
00:08:38.669 --> 00:08:43.549
AI as an AI is for my company, Libora, is sending
00:08:43.549 --> 00:08:47.590
the cold emails. So what I did is I created a
00:08:47.590 --> 00:08:52.389
system where I have the contacts of the potential
00:08:52.389 --> 00:08:56.789
clients. Those clients, I created an email list.
00:08:57.210 --> 00:09:01.090
where the AI is going to read the name, the address,
00:09:01.250 --> 00:09:04.389
and the website of the clinic, and they will
00:09:04.389 --> 00:09:08.490
customize the text of that email. It will customize
00:09:08.490 --> 00:09:10.850
the text of the email. It will send through and
00:09:10.850 --> 00:09:13.409
attach it with another software. It will send
00:09:13.409 --> 00:09:16.529
my first email as that, okay, I'm Charmaine,
00:09:16.610 --> 00:09:19.929
and I see that your clinic based in this area.
00:09:21.149 --> 00:09:23.269
If you are looking for funding, it will be there.
00:09:23.350 --> 00:09:26.929
So first email, second email as a follow -up.
00:09:27.200 --> 00:09:30.059
And third as a last email as a third follow up.
00:09:30.200 --> 00:09:33.740
And so I'm sending those package of emails like
00:09:33.740 --> 00:09:37.200
10 to 25 times a day. And again, what I have
00:09:37.200 --> 00:09:39.639
to do is to just look at what's going on, click
00:09:39.639 --> 00:09:42.340
OK, instead of writing them all. Wow. That is
00:09:42.340 --> 00:09:45.159
a lot of research to have done on your behalf.
00:09:45.419 --> 00:09:48.039
Yeah. You know what? It seems like now it's automatic,
00:09:48.259 --> 00:09:50.820
but there is hours and hours of planning and
00:09:50.820 --> 00:09:53.740
how to solve it and how to do it. Finally, we're
00:09:53.740 --> 00:09:55.980
happy that we created this system where we can
00:09:55.980 --> 00:09:59.580
generate emails and send them automatically once
00:09:59.580 --> 00:10:03.279
the system will stop if they reply. And you mentioned
00:10:03.279 --> 00:10:06.779
that the problem you're solving is matching up
00:10:06.779 --> 00:10:10.480
clinics, lenders. So they want to borrow some
00:10:10.480 --> 00:10:13.679
amount of money. What is the actual problem?
00:10:13.960 --> 00:10:15.360
What are the clinics looking to borrow money
00:10:15.360 --> 00:10:19.360
for? And how did you come across this mismatch
00:10:19.360 --> 00:10:22.690
where people find it hard to borrow? So it started
00:10:22.690 --> 00:10:24.809
with the casual discussions that I was having
00:10:24.809 --> 00:10:28.909
with my massages, with the dental clinic. And
00:10:28.909 --> 00:10:31.850
they kept coming, saying the one thing repetitively
00:10:31.850 --> 00:10:34.629
is that they had a growing business, but they're
00:10:34.629 --> 00:10:38.049
having a hard time to get loans for their business.
00:10:38.690 --> 00:10:41.090
Sometimes they have to put their own house as
00:10:41.090 --> 00:10:43.950
a collateral to get the loans. When I started
00:10:43.950 --> 00:10:46.909
researching, I... the research was showing me
00:10:46.909 --> 00:10:49.029
different things and the reality was showing
00:10:49.029 --> 00:10:51.529
me the different thing on the paper on the website
00:10:51.529 --> 00:10:54.090
they say all you need to have is six months in
00:10:54.090 --> 00:10:58.110
business at least 15 to 20k in in revenues per
00:10:58.110 --> 00:11:01.190
month and then what is the third thing yeah your
00:11:01.190 --> 00:11:04.340
bank statements that's it But on the other side,
00:11:04.460 --> 00:11:07.639
the borrowers or the potential borrowers, they
00:11:07.639 --> 00:11:10.220
were having a hard time. So this is where I found
00:11:10.220 --> 00:11:12.820
the gap that the information is there, the lenders
00:11:12.820 --> 00:11:15.559
are there, but the accessibility is difficult.
00:11:15.820 --> 00:11:18.259
And so this is where I came in, is that, okay,
00:11:18.360 --> 00:11:21.340
through Web2, through Levora, if you apply with
00:11:21.340 --> 00:11:23.840
Levora, I will match you with multiple lenders.
00:11:23.960 --> 00:11:27.080
You don't have to go each lender one by one and
00:11:27.080 --> 00:11:29.299
compare their offer. And it's going to be all
00:11:29.299 --> 00:11:33.009
online. It's a brokerage platform online. But
00:11:33.009 --> 00:11:35.970
the only difference is that instead of giving
00:11:35.970 --> 00:11:39.429
one best offer, we will recommend an offer through
00:11:39.429 --> 00:11:42.850
my AI, but we will still show them the other
00:11:42.850 --> 00:11:45.269
offers that other lenders are giving. So the
00:11:45.269 --> 00:11:48.649
borrower has this spectrum of offers and choose
00:11:48.649 --> 00:11:52.049
the one which fits them best. I find it fascinating
00:11:52.049 --> 00:11:55.570
as well that it isn't related to your day job.
00:11:55.710 --> 00:11:58.649
This doesn't seem like a problem that you came
00:11:58.649 --> 00:12:01.169
up through that industry. How do you feel about
00:12:01.169 --> 00:12:03.149
encountering this problem and then how did you
00:12:03.149 --> 00:12:05.590
make that decision to jump in and solve it? I
00:12:05.590 --> 00:12:08.710
discovered so many things while being an entrepreneur
00:12:08.710 --> 00:12:11.769
in this journey. I thought that if I solve the
00:12:11.769 --> 00:12:14.389
human -centered design problem, if I will be
00:12:14.389 --> 00:12:17.070
able to match the borrower with the lenders,
00:12:17.169 --> 00:12:19.730
I'm going to save their time with one streamlined
00:12:19.730 --> 00:12:23.559
application. they will be able to see multiple
00:12:23.559 --> 00:12:26.299
offers. It's all user experience design that
00:12:26.299 --> 00:12:28.799
I was working on. So my initial idea that I'm
00:12:28.799 --> 00:12:30.940
going to use my designing skill to make it a
00:12:30.940 --> 00:12:34.019
better user experience. But in reality, it's
00:12:34.019 --> 00:12:38.080
only 10 % of the work. The other 90 % things
00:12:38.080 --> 00:12:42.159
are not design related. And so, yeah, I underestimated
00:12:42.159 --> 00:12:44.580
the amount of work that a founder needs to do.
00:12:44.639 --> 00:12:47.980
And yes, my background helped me a little bit,
00:12:48.080 --> 00:12:50.929
but I'm still learning. There are hundreds of
00:12:50.929 --> 00:12:54.149
other wheels which need to move all together
00:12:54.149 --> 00:12:57.129
to make a meaningful product, to create a meaningful
00:12:57.129 --> 00:13:00.450
company. And yes, my background helped me for
00:13:00.450 --> 00:13:03.169
making things user -friendly, but not more than
00:13:03.169 --> 00:13:08.289
that. And before LiveAura existed, what was the
00:13:08.289 --> 00:13:11.210
experience of both the borrower and the lender?
00:13:11.309 --> 00:13:15.519
How does that improve now? So we still have a
00:13:15.519 --> 00:13:17.539
lot of work to do because we haven't reached
00:13:17.539 --> 00:13:21.759
the main population that we want to. So it's
00:13:21.759 --> 00:13:24.000
not that they had this experience and now it's
00:13:24.000 --> 00:13:27.080
completed. This problem is still there. The number
00:13:27.080 --> 00:13:31.100
one problem is whenever they get rejected by
00:13:31.100 --> 00:13:34.360
a bank, they get frustrated and they feel that,
00:13:34.399 --> 00:13:37.840
okay, that's it. So most of the... Borrowers,
00:13:37.919 --> 00:13:40.279
they don't even apply. And according to the Federal
00:13:40.279 --> 00:13:44.539
Reserve Statistic of 2025, 60 % of the people,
00:13:44.620 --> 00:13:48.200
they don't reapply to other banks. And nobody
00:13:48.200 --> 00:13:50.059
is there who can tell me that, okay, this is
00:13:50.059 --> 00:13:51.620
not the end of the world. There are other people
00:13:51.620 --> 00:13:53.820
you might get approved. They think that everyone
00:13:53.820 --> 00:13:57.059
is going to be the same. And so this is the biggest
00:13:57.059 --> 00:14:00.500
thing. The educational part is still what we
00:14:00.500 --> 00:14:04.049
are working on to tell borrowers that. If you
00:14:04.049 --> 00:14:06.690
got rejected by your bank, it doesn't mean that
00:14:06.690 --> 00:14:09.769
you will be rejected by others as well. And what
00:14:09.769 --> 00:14:12.710
other things we are doing is that maybe today
00:14:12.710 --> 00:14:16.309
they cannot get a loan because they have something
00:14:16.309 --> 00:14:19.690
which is missing. And so that's why we also developed
00:14:19.690 --> 00:14:23.149
a borrower readiness portal where we will ask
00:14:23.149 --> 00:14:26.330
a couple of questions. For example, how much
00:14:26.330 --> 00:14:28.870
revenues you have? What is your profit margin?
00:14:29.250 --> 00:14:32.990
How much documents you have today? and we give
00:14:32.990 --> 00:14:36.129
them a score that okay you have 80 of chances
00:14:36.129 --> 00:14:40.509
to be to qualify in other areas and some if they
00:14:40.509 --> 00:14:43.190
have something missing or if they are not eligible
00:14:43.190 --> 00:14:47.789
today we also help them by giving them some spreadsheet
00:14:47.789 --> 00:14:50.710
to track their their revenues and how to create
00:14:50.710 --> 00:14:54.009
the forecast so that they will be eligible tomorrow
00:14:54.009 --> 00:14:58.490
so our aim is not only to streamline the application
00:14:58.490 --> 00:15:01.779
but to make the whole experience From finding
00:15:01.779 --> 00:15:05.019
a loan, if you get the loan with multiple offer,
00:15:05.179 --> 00:15:07.679
if you cannot get the loan today, how can we
00:15:07.679 --> 00:15:10.000
help you to prepare tomorrow to get the loan?
00:15:10.159 --> 00:15:13.399
So all those things are what we are working today.
00:15:13.820 --> 00:15:16.639
There are two angles here that feel there's a
00:15:16.639 --> 00:15:19.879
lot to unpack. And one of them is, how do the
00:15:19.879 --> 00:15:22.340
lenders feel about this platform? And the other
00:15:22.340 --> 00:15:25.379
is, I'm interested in what role AI plays, as
00:15:25.379 --> 00:15:29.629
well as what can just be satisfied with. other
00:15:29.629 --> 00:15:32.090
forms of technology, maybe really dumb forms
00:15:32.090 --> 00:15:35.470
of technology as well. So Daniel, there's one
00:15:35.470 --> 00:15:38.409
thing that I don't see AI is playing any role,
00:15:38.570 --> 00:15:43.110
building trust. So I can crush that trust with
00:15:43.110 --> 00:15:47.289
the AI. So this is where I'm avoiding AI. I'm
00:15:47.289 --> 00:15:49.730
giving you one example. Whenever I speak to my
00:15:49.730 --> 00:15:53.090
lenders, it's me and them. There is no one who's
00:15:53.090 --> 00:15:55.169
giving me script that this is what you should
00:15:55.169 --> 00:15:58.370
be speaking. No, it's my thoughts. It's my language.
00:15:59.210 --> 00:16:02.190
And yes, I can generate the content, maybe give
00:16:02.190 --> 00:16:04.950
me the topics that I can give you. These are
00:16:04.950 --> 00:16:06.830
the topics you should speak to them. But when
00:16:06.830 --> 00:16:08.649
I'm building them, they are trusting me. They're
00:16:08.649 --> 00:16:12.409
meeting a human. Another way is also creating
00:16:12.409 --> 00:16:16.610
the trust with my borrowers. I started producing
00:16:16.610 --> 00:16:20.429
content, educational content on LinkedIn and
00:16:20.429 --> 00:16:24.200
now recently started also on Instagram. I'm using
00:16:24.200 --> 00:16:27.820
AI to refine my ideas. But where I don't use
00:16:27.820 --> 00:16:31.100
AI is to, okay, AI, create an image for me and
00:16:31.100 --> 00:16:32.960
just give me a topic and then just copy paste
00:16:32.960 --> 00:16:37.620
it in there. So AI, where it can be a huge help
00:16:37.620 --> 00:16:41.440
to build product to go faster, but it can also
00:16:41.440 --> 00:16:44.539
destroy your identity if they find you spam AI,
00:16:44.779 --> 00:16:47.840
like, okay, I'm just clicking and doing it. So
00:16:47.840 --> 00:16:51.009
using AI, you need to understand. How you're
00:16:51.009 --> 00:16:53.590
using it. Are you using it to build your image
00:16:53.590 --> 00:16:56.850
or destroy your image? I like that people trust
00:16:56.850 --> 00:16:59.309
humans. They don't trust AI. Yeah. And especially
00:16:59.309 --> 00:17:02.870
the founder. Whenever I call, so I started also
00:17:02.870 --> 00:17:04.789
cold calling. I said, this is Sherman, founder
00:17:04.789 --> 00:17:07.329
of Levora. It's hard for me still to imagine
00:17:07.329 --> 00:17:10.849
that an AI will be speaking from my side to say
00:17:10.849 --> 00:17:12.490
that, hi, this is Sherman, founder of Levora.
00:17:12.710 --> 00:17:15.410
Yeah, it's because I'm speaking human to human.
00:17:17.109 --> 00:17:20.309
On the other hand, I am using AI to generate
00:17:20.309 --> 00:17:23.269
emails because this is where I've got the contact
00:17:23.269 --> 00:17:25.650
and I said that, okay, these are the contacts,
00:17:25.710 --> 00:17:28.710
send me my value proposition. I'm sending them
00:17:28.710 --> 00:17:31.650
a value proposition. But the moment they click
00:17:31.650 --> 00:17:34.170
the button, book an appointment, this is now
00:17:34.170 --> 00:17:36.150
I see that they're interested and now they see
00:17:36.150 --> 00:17:40.150
me. So yes, AI has their limit. I am using it
00:17:40.150 --> 00:17:42.589
in many areas, but it's not that if you click
00:17:42.589 --> 00:17:44.390
on the book meeting and there will be another
00:17:44.390 --> 00:17:47.769
chatbot who's going to speak with you. And as
00:17:47.769 --> 00:17:50.630
far as the product that you've built goes, how
00:17:50.630 --> 00:17:53.269
does the divide between AI and technology look
00:17:53.269 --> 00:17:57.609
there? One thing is I am not a software engineer.
00:17:58.170 --> 00:18:01.730
So initially, it was harder for me to estimate
00:18:01.730 --> 00:18:04.109
how much time it will take to build my product.
00:18:04.369 --> 00:18:07.150
But I had been working in a company, so I had
00:18:07.150 --> 00:18:09.569
a rough estimate. And this is where I shocked
00:18:09.569 --> 00:18:12.269
the impact of an AI, especially for the new generation.
00:18:12.890 --> 00:18:15.910
I'm working with the students. And initially,
00:18:16.029 --> 00:18:19.349
I estimated that it will take them 300 hours
00:18:19.349 --> 00:18:22.630
to create the first portal. And when I showed
00:18:22.630 --> 00:18:25.130
them the portal and asked them that, OK, this
00:18:25.130 --> 00:18:27.670
is the workflow. These are the prototypes. And
00:18:27.670 --> 00:18:30.269
they understood the project itself. And they
00:18:30.269 --> 00:18:33.250
gave me an estimate of 60 hours instead of 300
00:18:33.250 --> 00:18:37.430
hours. It's because today's generation who never
00:18:37.430 --> 00:18:40.569
learned or maybe learned by they were in school
00:18:40.569 --> 00:18:43.349
a little bit how to generate code without AI.
00:18:43.819 --> 00:18:46.660
But now they are developing their skill with
00:18:46.660 --> 00:18:49.500
AI. It's not something like people who had been
00:18:49.500 --> 00:18:52.539
a developer for the last 20, 30 years and now
00:18:52.539 --> 00:18:55.599
they have to relearn. Now this is a completely
00:18:55.599 --> 00:18:58.680
different game. They are learning with AI. So
00:18:58.680 --> 00:19:01.380
they are efficient from the beginning. They don't
00:19:01.380 --> 00:19:03.920
know any other way to do it. So this is where
00:19:03.920 --> 00:19:07.380
I was impressed that how AI is helping software
00:19:07.380 --> 00:19:12.099
developers to go faster. And inside of... product
00:19:12.099 --> 00:19:15.460
features. You've mentioned that there is some
00:19:15.460 --> 00:19:18.440
analysis of lenders. There is some matching.
00:19:18.680 --> 00:19:21.140
I think you said analysis of borrowers as well
00:19:21.140 --> 00:19:24.160
that does have an AI component. Yes. Is that
00:19:24.160 --> 00:19:27.660
correct? And is there more? There is an AI component
00:19:27.660 --> 00:19:31.819
because whenever the borrower, they give all
00:19:31.819 --> 00:19:35.279
their information, the AI helps me to understand
00:19:35.279 --> 00:19:38.319
that if they are ready to borrow or not. So this
00:19:38.319 --> 00:19:42.319
system is all AI. And also, This is the first
00:19:42.319 --> 00:19:43.900
stage when they say they're ready or not. But
00:19:43.900 --> 00:19:46.019
the second stage, okay, they are ready, but who
00:19:46.019 --> 00:19:48.400
is going to be the best match? Again, the AI
00:19:48.400 --> 00:19:50.740
is going to match that who's going to be the
00:19:50.740 --> 00:19:53.859
best lender for them. If you're at liberty to
00:19:53.859 --> 00:19:58.059
share, how do you figure out a lender would be
00:19:58.059 --> 00:20:00.920
a good match for someone? Is that publicly available
00:20:00.920 --> 00:20:03.680
information? Is that based on history or what?
00:20:04.039 --> 00:20:07.700
It can be. For some lenders, it is publicly available
00:20:07.700 --> 00:20:11.190
information. giving you an example merchant growth
00:20:11.190 --> 00:20:13.890
they are one of the biggest lenders of canada
00:20:13.890 --> 00:20:16.670
they can go and you can see in their website
00:20:16.670 --> 00:20:21.390
what are the main criteria but the thing is today
00:20:21.390 --> 00:20:23.769
as i still told you that when borrowers they
00:20:23.769 --> 00:20:26.950
get rejected They don't even go in deep and see
00:20:26.950 --> 00:20:29.369
that, am I going to be eligible for this or that?
00:20:29.470 --> 00:20:31.569
They just don't go there. There's a psychological
00:20:31.569 --> 00:20:35.549
barrier. So for so many lenders, it is publicly
00:20:35.549 --> 00:20:38.789
available. One in the United States, our partner
00:20:38.789 --> 00:20:42.950
is AFN, America Funding Network. They also have
00:20:42.950 --> 00:20:45.829
their information available. But the thing is,
00:20:45.910 --> 00:20:49.049
sometimes those information are not just like
00:20:49.049 --> 00:20:52.349
a landing page. There are five videos that you
00:20:52.349 --> 00:20:54.910
can watch and then you can see that, okay. So
00:20:54.910 --> 00:20:58.049
yes, when you, again, when you're running a clinic,
00:20:58.230 --> 00:21:01.009
you have other things to do than watching those
00:21:01.009 --> 00:21:03.869
videos. And so like me, like you said earlier,
00:21:03.990 --> 00:21:05.970
that seven hours doesn't look a lot, but we have
00:21:05.970 --> 00:21:08.150
a lot of things to do. So should I spend my time
00:21:08.150 --> 00:21:12.670
looking at that or just not? So in some cases,
00:21:12.769 --> 00:21:14.970
the information is available, but it's hard to
00:21:14.970 --> 00:21:17.849
find. It's hard to access. And even if it's easy
00:21:17.849 --> 00:21:21.410
to access, they don't go and look at. And that
00:21:21.410 --> 00:21:24.279
way. You've already pre -processed all this information
00:21:24.279 --> 00:21:27.259
and are able to make that judgment across many
00:21:27.259 --> 00:21:29.940
lenders. Yes, and sometimes they don't even need
00:21:29.940 --> 00:21:32.740
to know what other lenders. They just apply.
00:21:33.240 --> 00:21:36.160
In the backend, I have 10 -ish lenders. I have
00:21:36.160 --> 00:21:38.480
all their information. I match them with them.
00:21:38.519 --> 00:21:41.299
I match the borrowers with them. And then they
00:21:41.299 --> 00:21:45.339
will just get the best offer or, sorry, multiple
00:21:45.339 --> 00:21:48.289
offers with the recommended offer. So they don't
00:21:48.289 --> 00:21:51.529
have to go and search that how much this lender
00:21:51.529 --> 00:21:54.049
is doing. I already have those information and
00:21:54.049 --> 00:21:57.589
it's already applied. And how does that affect
00:21:57.589 --> 00:22:01.470
the lenders? Lenders, building the company came
00:22:01.470 --> 00:22:04.069
up with so many good and bad surprises. This
00:22:04.069 --> 00:22:06.750
is actually a good surprise. I was extremely
00:22:06.750 --> 00:22:10.170
shy when I was speaking to my first lending partner.
00:22:10.490 --> 00:22:14.140
I was this imposter syndrome to say that. Someone
00:22:14.140 --> 00:22:16.839
who's not from finance, not from mortgage, a
00:22:16.839 --> 00:22:19.480
designer, how she's coming and developing this
00:22:19.480 --> 00:22:22.680
platform. But my initial discussion, all my discussion
00:22:22.680 --> 00:22:26.059
had been extremely smooth. And I came to know
00:22:26.059 --> 00:22:29.839
that what lender wants is qualified borrowers.
00:22:30.140 --> 00:22:33.119
And this is one thing that what I wanted. And
00:22:33.119 --> 00:22:36.779
this is what All Libora was all about. So we
00:22:36.779 --> 00:22:39.079
are not just a website where people can just
00:22:39.079 --> 00:22:41.619
click and then I say that, OK. there are these
00:22:41.619 --> 00:22:44.099
three companies clicked on my website you can
00:22:44.099 --> 00:22:49.240
see that i actually filter the qualified borrowers
00:22:49.240 --> 00:22:52.400
pre -screen them and then send it to them the
00:22:52.400 --> 00:22:54.500
the final decision is going to be always the
00:22:54.500 --> 00:22:57.500
lender if they are okay or not but at least they
00:22:57.500 --> 00:23:01.299
are i'm not giving them a huge quantity of borrowers
00:23:01.299 --> 00:23:03.799
and then they have to filter to find one i'm
00:23:03.799 --> 00:23:06.039
already giving them someone who are already pre
00:23:06.039 --> 00:23:08.079
-screened who already have their bank statements
00:23:08.079 --> 00:23:11.720
their number of in business so it's a huge time
00:23:11.720 --> 00:23:15.420
saving for lenders and also they would love to
00:23:15.420 --> 00:23:18.420
help you they would love to give loans but this
00:23:18.420 --> 00:23:22.160
is how they work And they welcomed me with whole
00:23:22.160 --> 00:23:25.420
hands that, OK, just be our partner. We'll be
00:23:25.420 --> 00:23:27.920
happy to help you as a partner. And do you need
00:23:27.920 --> 00:23:31.920
to adjust for each lender? Or does each lender
00:23:31.920 --> 00:23:34.420
then say, oh, these leads are coming from Livora.
00:23:34.859 --> 00:23:37.579
That means they're high quality. How does that
00:23:37.579 --> 00:23:40.400
relationship evolve? There was a small adjustment
00:23:40.400 --> 00:23:43.680
because there are a few lenders who don't like.
00:23:44.379 --> 00:23:47.259
that I'm also sending other lenders the same
00:23:47.259 --> 00:23:49.579
information and the borrower will choose the
00:23:49.579 --> 00:23:51.859
best. Upfront, they said that, you know what,
00:23:51.960 --> 00:23:54.900
if we want it, we want it. And we don't want
00:23:54.900 --> 00:23:57.579
to compete with other lenders. So at that time,
00:23:57.759 --> 00:24:00.119
why would they want to compete? Someone else
00:24:00.119 --> 00:24:02.920
will take it. So yeah, and this is where I said
00:24:02.920 --> 00:24:06.059
that maybe I created their career board separately.
00:24:06.680 --> 00:24:09.460
This is not the foundation of Livora. We are
00:24:09.460 --> 00:24:12.539
here for transparency. We are here to match with
00:24:12.539 --> 00:24:16.019
borrower. If I find someone and it does not apply
00:24:16.019 --> 00:24:18.039
to any other lender, then I will be happy to
00:24:18.039 --> 00:24:21.420
show you and if you qualify for it. Okay. That
00:24:21.420 --> 00:24:23.420
is, I think, one of the interesting consequences
00:24:23.420 --> 00:24:28.740
of being able to pre -screen to qualify people
00:24:28.740 --> 00:24:32.259
and look across so many lenders is now you've
00:24:32.259 --> 00:24:35.599
created essentially a marketplace where the lenders
00:24:35.599 --> 00:24:37.779
have to compete against each other. Exactly.
00:24:37.920 --> 00:24:40.259
Yeah, this is kind of a marketplace. This is
00:24:40.259 --> 00:24:42.819
a very good word that you just said. It's because
00:24:42.819 --> 00:24:47.380
we are not direct lenders. We are not unlocking
00:24:47.380 --> 00:24:50.359
any money or that. We are just a marketplace
00:24:50.359 --> 00:24:54.240
where we will match them. And by the way, there's
00:24:54.240 --> 00:24:57.339
a small story. Last year when I also wanted to
00:24:57.339 --> 00:25:00.019
become going in this is I was interviewed by
00:25:00.019 --> 00:25:03.240
a company called Pine. It's one of the fastest
00:25:03.240 --> 00:25:06.819
growing startup in Canada. And they streamlined
00:25:06.819 --> 00:25:09.259
the process for mortgages. So I was speaking
00:25:09.259 --> 00:25:12.839
with the CEO of that startup and he told me that
00:25:12.839 --> 00:25:15.619
today there is a lot of potential of creating
00:25:15.619 --> 00:25:19.500
those kind of marketplaces. There are so many
00:25:19.500 --> 00:25:22.200
areas that the people did not even touch and
00:25:22.200 --> 00:25:25.019
lending platform is one of them, especially in
00:25:25.019 --> 00:25:27.720
North America. So there are two giants in the
00:25:27.720 --> 00:25:31.880
U .S., but they still need more work, more specialized
00:25:31.880 --> 00:25:34.339
people who can go in and make it even better.
00:25:34.660 --> 00:25:38.160
And as you develop this, As you develop this
00:25:38.160 --> 00:25:40.619
marketplace, what's maybe one of the biggest
00:25:40.619 --> 00:25:43.000
positive surprises and one of the biggest negative
00:25:43.000 --> 00:25:46.460
surprises? The negative surprises for me is the
00:25:46.460 --> 00:25:50.779
hustle of communication. Where to find the borrowers?
00:25:50.880 --> 00:25:54.160
There are people who need it, but where are they?
00:25:54.380 --> 00:25:59.680
So marketplace, it comes with also, it's an online
00:25:59.680 --> 00:26:02.880
thing. It's an online app. And then again, the
00:26:02.880 --> 00:26:05.769
trust issue comes. why would they maybe it's
00:26:05.769 --> 00:26:08.589
an ai thing and there's no human behind it so
00:26:08.589 --> 00:26:12.890
how someone is sitting in florida and he can
00:26:12.890 --> 00:26:15.970
say that okay i trust this platform and i'm okay
00:26:15.970 --> 00:26:20.150
to share that to ask it so yes this is one of
00:26:20.150 --> 00:26:23.569
the thing that is a biggest hustle for me that
00:26:23.569 --> 00:26:26.470
how can i show them that i'm a trustworthy person
00:26:26.470 --> 00:26:28.849
and in order to do that i have to communicate
00:26:28.849 --> 00:26:32.430
i have to create YouTube videos, and somehow
00:26:32.430 --> 00:26:35.269
I reach them and to say that, yes, you have this
00:26:35.269 --> 00:26:37.869
problem. I am one of the person who can solve
00:26:37.869 --> 00:26:40.950
it. So building trust and domain authority. Yes.
00:26:41.490 --> 00:26:45.289
Also the reach is how can I reach those people
00:26:45.289 --> 00:26:49.329
who need it? But yeah, where are they? Still,
00:26:49.650 --> 00:26:52.210
I'm figuring out how to do that. And what about
00:26:52.210 --> 00:26:55.579
a positive surprise? The biggest... positive
00:26:55.579 --> 00:26:58.779
surprise is i thought that i shared it a little
00:26:58.779 --> 00:27:00.900
bit but i would say that is the most positive
00:27:00.900 --> 00:27:04.920
surprise is lenders trusted me and my platform
00:27:04.920 --> 00:27:08.279
and they became my partners so it was piece of
00:27:08.279 --> 00:27:11.160
cake for how to make them and i just made not
00:27:11.160 --> 00:27:13.839
one like five ten it's just like this and so
00:27:13.839 --> 00:27:15.500
i thought that this is going to be the biggest
00:27:15.500 --> 00:27:18.259
hustle but no actually apparently there are two
00:27:18.259 --> 00:27:20.940
side of network and livor it's all about two
00:27:20.940 --> 00:27:25.299
-sided network lenders and borrowers the Pleasant
00:27:25.299 --> 00:27:28.500
surprise was how easy it was to create the network
00:27:28.500 --> 00:27:31.720
of lenders. And the hustle, the bad surprise
00:27:31.720 --> 00:27:34.579
was how hard it is going to be to create a network
00:27:34.579 --> 00:27:38.519
of borrowers. That makes sense. But I also am
00:27:38.519 --> 00:27:40.619
a little surprised that lenders are so eager
00:27:40.619 --> 00:27:43.539
for this, but that's great. One of the things
00:27:43.539 --> 00:27:47.880
I find fascinating about AI is it allows these
00:27:47.880 --> 00:27:49.740
kind of connections that would have taken a lot
00:27:49.740 --> 00:27:52.640
of human judgment, but are so necessary. On one
00:27:52.640 --> 00:27:54.480
side, they're discouraging when it doesn't go
00:27:54.480 --> 00:27:56.960
right. On the other side, it's a ton of money
00:27:56.960 --> 00:28:00.099
to invest in examining every single borrower.
00:28:00.380 --> 00:28:03.420
But now AI has lowered this barrier, making that
00:28:03.420 --> 00:28:08.680
connection through Libora. Yes, and AI is, I'm
00:28:08.680 --> 00:28:12.240
so surprised that when I started AI with the
00:28:12.240 --> 00:28:16.059
chat GPT in late 2022, or I don't remember, maybe
00:28:16.059 --> 00:28:18.579
in late 2022, I thought that it's going to just,
00:28:18.660 --> 00:28:20.640
I ask a question, it will give me an answer.
00:28:21.160 --> 00:28:25.079
But today, AI is, we are using it in so many
00:28:25.079 --> 00:28:27.160
different angles that I never thought that AI
00:28:27.160 --> 00:28:30.940
might be useful for that. And this angle of matching
00:28:30.940 --> 00:28:33.920
borrowers with the lenders, this is one of the
00:28:33.920 --> 00:28:36.559
biggest angles that I'm using. Apart from building
00:28:36.559 --> 00:28:39.079
my own company where I can generate emails, send
00:28:39.079 --> 00:28:41.680
it to them, or the software which can be developed
00:28:41.680 --> 00:28:45.200
10 times faster than it used to be. Also, new
00:28:45.200 --> 00:28:48.279
things that we are discovering which can be done
00:28:48.279 --> 00:28:51.640
with the AI. wouldn't have been possible before.
00:28:53.359 --> 00:28:57.799
Are there any new possibilities that are exciting
00:28:57.799 --> 00:28:59.480
to you that wouldn't have been possible before?
00:28:59.859 --> 00:29:02.460
You know what? I'm still learning AI. Every day
00:29:02.460 --> 00:29:05.079
is a different day for AI. Give you an example.
00:29:05.460 --> 00:29:09.359
Yesterday, I was looking at the Excel sheet and
00:29:09.359 --> 00:29:11.859
I was not understanding what is going on in here.
00:29:12.039 --> 00:29:15.579
What I did, I just simply created a GPT. and
00:29:15.579 --> 00:29:17.700
say that, okay, I'm going to now share with you
00:29:17.700 --> 00:29:21.140
some complex. What I want is, I want to use an
00:29:21.140 --> 00:29:23.759
easy English to deconstruct the whole spreadsheet
00:29:23.759 --> 00:29:26.599
and etc. And then it just did it. Otherwise,
00:29:26.700 --> 00:29:29.460
I just spend my time on analyzing it by myself.
00:29:29.720 --> 00:29:32.000
All I, now all I have to do is just drag and
00:29:32.000 --> 00:29:34.500
drop and say, okay, explain me how to deconstruct
00:29:34.500 --> 00:29:36.579
it. It seems like a simple thing that anyone
00:29:36.579 --> 00:29:39.400
can do, but unless you haven't done it, you don't
00:29:39.400 --> 00:29:41.519
even think about it that you can do it with the
00:29:41.519 --> 00:29:44.859
AI. Yeah, spreadsheet analysis is... It has come
00:29:44.859 --> 00:29:47.299
a long way. So I think maybe even last year,
00:29:47.339 --> 00:29:49.779
it wouldn't have worked that easily. And also
00:29:49.779 --> 00:29:54.420
the combination of multiple AI is I use perplexity
00:29:54.420 --> 00:29:56.859
for a different reason. I'm using Grok more than
00:29:56.859 --> 00:30:02.079
perplexity. So Grok with the chat GPT, Cloud
00:30:02.079 --> 00:30:05.480
I used, but I'm going to switch with Cloud again.
00:30:05.960 --> 00:30:09.930
And then... The UX pilot, it's another AI. So
00:30:09.930 --> 00:30:12.509
I'm using all those combinations of different
00:30:12.509 --> 00:30:16.190
AI to create one solution. That's cool. What
00:30:16.190 --> 00:30:18.190
do you like about each one of those solutions?
00:30:18.670 --> 00:30:22.410
What I love about Grok is that the instant information.
00:30:23.109 --> 00:30:26.609
Something happened three hours ago. I ask a question
00:30:26.609 --> 00:30:29.029
that, okay, what's the most exciting question
00:30:29.029 --> 00:30:31.730
about lending in North America, in Canada, and
00:30:31.730 --> 00:30:33.910
in the U .S.? And then it searches and gives
00:30:33.910 --> 00:30:37.500
me the latest news. about it so i'm not i'm not
00:30:37.500 --> 00:30:41.440
sure if chat gpt is as quick as not it's also
00:30:41.440 --> 00:30:45.940
because maybe the grok is made by x twitter and
00:30:45.940 --> 00:30:47.980
everything which is in twitter it's already in
00:30:47.980 --> 00:30:50.299
their platform and then you just access it quickly
00:30:50.299 --> 00:30:54.000
so grok had been it had been a great tool for
00:30:54.000 --> 00:30:57.619
me get this access of the information chat gpt
00:30:57.619 --> 00:31:01.700
is my go -to thing go to ai it just got so it's
00:31:01.700 --> 00:31:05.779
so easy to use i created my folders And I created
00:31:05.779 --> 00:31:09.039
GPTs inside. Maybe today we are saying, and I
00:31:09.039 --> 00:31:11.859
also create my LinkedIn post because I have GPTs,
00:31:11.920 --> 00:31:14.920
which helps me to create LinkedIn posts. Apparently,
00:31:14.920 --> 00:31:18.940
I'm hearing that Claude is getting better in
00:31:18.940 --> 00:31:23.039
creating your sound and your text better than
00:31:23.039 --> 00:31:25.720
GPT. I haven't tested Claude yet, but so far
00:31:25.720 --> 00:31:28.779
GPT is my go -to tool for all the other things.
00:31:28.960 --> 00:31:32.099
It also has all my history where I started. Yeah,
00:31:32.119 --> 00:31:36.089
this is. And there is. Third AI which literally
00:31:36.089 --> 00:31:39.910
changed my life after using AI. WhisperFlow.
00:31:41.369 --> 00:31:44.869
WhisperFlow, I started using. I don't type anymore
00:31:44.869 --> 00:31:48.730
anything. So I just, if I need to write an email,
00:31:48.809 --> 00:31:52.369
I press a button and I speak. And when I stop
00:31:52.369 --> 00:31:55.170
it, it's just all written exactly the way I said.
00:31:55.309 --> 00:31:57.609
All the capitals are there, all the commas, all
00:31:57.609 --> 00:32:01.029
the stops are there. And if you start using WhisperFlow,
00:32:01.170 --> 00:32:04.230
I highly recommend to test it. It will save a
00:32:04.230 --> 00:32:07.569
ton of your time. And I think in a few years,
00:32:07.630 --> 00:32:10.509
the keyboard is going to disappear if we have
00:32:10.509 --> 00:32:14.089
tools like WhisperFlow. I like that. I can keep
00:32:14.089 --> 00:32:17.470
going. There are so many. But I would say top
00:32:17.470 --> 00:32:20.869
three, GPT, Grok, and WhisperFlow. No, thank
00:32:20.869 --> 00:32:24.190
you. I think like you have alluded to, it's quite
00:32:24.190 --> 00:32:26.809
tough to keep up with all the changes. So hearing
00:32:26.809 --> 00:32:28.829
how people are currently using them and how they've
00:32:28.829 --> 00:32:31.869
enjoyed them is very helpful. So thank you. No,
00:32:31.950 --> 00:32:34.890
I'm also getting updated every day. It's quite
00:32:34.890 --> 00:32:38.650
overwhelming, things which are progressing. So
00:32:38.650 --> 00:32:41.410
maybe what I said today, it wouldn't be valid
00:32:41.410 --> 00:32:44.829
three months later. Yeah, that's quite possible.
00:32:45.170 --> 00:32:48.809
And I was interested in hearing from that angle,
00:32:48.970 --> 00:32:52.970
how has what you've done with Libora and being
00:32:52.970 --> 00:32:54.970
a mother at the same time as an entrepreneur,
00:32:55.230 --> 00:32:58.369
how has that been enabled by the way that AI
00:32:58.369 --> 00:33:01.180
is now capable? and maybe wasn't a year or two
00:33:01.180 --> 00:33:04.799
ago? First of all, it would be harder for me
00:33:04.799 --> 00:33:08.859
to take a decision to create my company. I was,
00:33:09.079 --> 00:33:13.920
maybe I, it's overwhelming being a mother. I
00:33:13.920 --> 00:33:16.940
don't have 18 hours a day to work for my company.
00:33:17.279 --> 00:33:19.599
I'm working like 18 hours a day, but it's not
00:33:19.599 --> 00:33:21.500
all for my company. The other half is for my
00:33:21.500 --> 00:33:24.779
kids, for my house. Again, this automation that
00:33:24.779 --> 00:33:28.019
is creating by the AI is just helping me to...
00:33:28.319 --> 00:33:33.380
Go faster, fail fast, reiterate fast. Being a
00:33:33.380 --> 00:33:36.039
mother, yes, I don't have this luxury of spending
00:33:36.039 --> 00:33:39.339
a lot of time only on my work. So making things
00:33:39.339 --> 00:33:43.660
faster and no -brainer things, those crafts where
00:33:43.660 --> 00:33:47.019
you don't need a human, AI is the answer for
00:33:47.019 --> 00:33:49.920
all those things. I like that. Do you have any
00:33:49.920 --> 00:33:52.519
specific examples of things that you don't think
00:33:52.519 --> 00:33:56.000
need a human? The best example is the writing.
00:33:56.589 --> 00:33:59.230
You don't need to, it's not that you don't need
00:33:59.230 --> 00:34:01.630
a human. You don't need your fingers to write
00:34:01.630 --> 00:34:04.390
the text. You can write it with your voice. You
00:34:04.390 --> 00:34:07.190
can do all the capitals and stops and commas
00:34:07.190 --> 00:34:09.449
and everything. You don't have to look on this.
00:34:09.989 --> 00:34:12.289
So I'm just saying that, oh my God, it's been
00:34:12.289 --> 00:34:15.010
so time. We started, we were thousands of years
00:34:15.010 --> 00:34:16.849
ago, we were writing with pencil and then there
00:34:16.849 --> 00:34:18.690
was a typewriter and then there's a keyboard
00:34:18.690 --> 00:34:22.400
and now no more fingers needed. So it's not about,
00:34:22.519 --> 00:34:25.780
yeah, human is, yeah, this is the one thing that
00:34:25.780 --> 00:34:27.880
I was extremely surprised that, oh my God, it's
00:34:27.880 --> 00:34:32.219
going to disappear in some time. And on the topic
00:34:32.219 --> 00:34:35.860
of not needing a human, has there been any or
00:34:35.860 --> 00:34:39.780
any errors that have come out of doing the analysis
00:34:39.780 --> 00:34:42.980
and matching borrowers with lenders in an automated
00:34:42.980 --> 00:34:47.340
way? Not yet. There are always errors because
00:34:47.340 --> 00:34:50.119
I am sitting behind and checking everything.
00:34:51.360 --> 00:34:54.179
I cannot leave everything on the AI to do it
00:34:54.179 --> 00:34:57.659
and just to send it. There are some small things,
00:34:57.699 --> 00:35:00.340
a small error somewhere AI does not understand.
00:35:00.460 --> 00:35:03.519
And it just translated in its own way and it
00:35:03.519 --> 00:35:06.159
comes back to something that you don't imagine.
00:35:06.460 --> 00:35:10.780
And so being a human is, at least as for today,
00:35:10.900 --> 00:35:15.039
in early 2026, you need a human behind who is
00:35:15.039 --> 00:35:17.059
just checking that everything is going smoothly.
00:35:17.639 --> 00:35:21.179
or if those are legitimate answers that you are
00:35:21.179 --> 00:35:23.280
expecting from the borrower. Today, we are not
00:35:23.280 --> 00:35:25.679
at a stage that only AI is doing everything.
00:35:25.980 --> 00:35:29.539
So do you foresee, at least for maybe the next
00:35:29.539 --> 00:35:33.000
six months, as Livora scales and handles more
00:35:33.000 --> 00:35:36.699
borrowers, you would need to hire people that
00:35:36.699 --> 00:35:39.619
would be behind the scenes, human in the loop
00:35:39.619 --> 00:35:43.119
style, and figuring out, should this be corrected?
00:35:43.440 --> 00:35:47.239
Yes. It's only if we have thousands of borrowers
00:35:47.239 --> 00:35:49.860
coming every second day. Maybe at that time it
00:35:49.860 --> 00:35:52.920
would be hard to scale with humans. But as for
00:35:52.920 --> 00:35:55.400
now, it's also we're in the early stage of the
00:35:55.400 --> 00:35:58.659
company. Trust is the thing that if we break
00:35:58.659 --> 00:36:02.619
this, we can just destroy our company from the
00:36:02.619 --> 00:36:05.860
beginning. And so for this trust, even if the
00:36:05.860 --> 00:36:10.539
AI is giving us best answers it can, we do need
00:36:10.539 --> 00:36:14.010
a human behind it. and it's not about only verifying
00:36:14.010 --> 00:36:15.929
the information or if the answer is correct it's
00:36:15.929 --> 00:36:18.309
also to understand okay this is the borrower
00:36:18.309 --> 00:36:21.130
they need it if they need a human connection
00:36:21.130 --> 00:36:24.130
behind it and for example recently i was speaking
00:36:24.130 --> 00:36:26.449
with someone who applied to my website and saw
00:36:26.449 --> 00:36:28.349
that how it was matched with the other lenders
00:36:28.349 --> 00:36:31.150
and but she still wanted to speak with someone
00:36:31.150 --> 00:36:35.000
as i had been following the whole process When
00:36:35.000 --> 00:36:37.539
I spoke to her, I knew exactly what's going on
00:36:37.539 --> 00:36:40.320
in her life and what are her questions. At this
00:36:40.320 --> 00:36:44.920
early stage, I believe that I myself has to be
00:36:44.920 --> 00:36:48.599
hands -on on all those areas. I like that. And
00:36:48.599 --> 00:36:51.099
that keeps the personal touch, even if there
00:36:51.099 --> 00:36:53.139
is an AI that's doing most of the grunt work.
00:36:53.599 --> 00:36:56.340
Yes, the personal touch and that trust, I would
00:36:56.340 --> 00:36:58.400
say. More than personal touch, this trust that
00:36:58.400 --> 00:37:01.539
there are humans behind the AI. It's not a spam
00:37:01.539 --> 00:37:05.269
AI. Yeah, that is important. And how do you see
00:37:05.269 --> 00:37:09.230
AI maybe changing in a way that would affect
00:37:09.230 --> 00:37:14.530
Libora or this marketplace business? Maybe, again,
00:37:14.710 --> 00:37:19.829
AI is going so fast that I will not be surprised
00:37:19.829 --> 00:37:23.210
that in five years, Libora, the whole proposition
00:37:23.210 --> 00:37:27.489
of one borrower matching with multiple lenders
00:37:27.489 --> 00:37:31.070
is going to be extremely... easy to use easy
00:37:31.070 --> 00:37:34.769
to develop and then there will be you i will
00:37:34.769 --> 00:37:38.010
no more need livora they will just go in a gpt
00:37:38.010 --> 00:37:42.250
apply and then they get the offers so maybe livora
00:37:42.250 --> 00:37:45.530
might not be existed at that time it's that's
00:37:45.530 --> 00:37:47.210
a very good question and this is what i was asking
00:37:47.210 --> 00:37:50.530
that maybe the ai is gonna replace livora maybe
00:37:51.190 --> 00:37:54.670
But one thing that I don't see AI doing is the
00:37:54.670 --> 00:37:57.590
network that I'm creating. It's the people that
00:37:57.590 --> 00:38:00.289
I'm connecting today and who knows me, who know
00:38:00.289 --> 00:38:04.030
what I do. Oh, yes, the application process might
00:38:04.030 --> 00:38:06.449
not be there, but the community I'm going to
00:38:06.449 --> 00:38:09.070
build from now until, I don't know, in five years.
00:38:09.230 --> 00:38:13.389
Today, there are hundreds of people in my connection
00:38:13.389 --> 00:38:15.750
of the borrower. But in five years, there will
00:38:15.750 --> 00:38:18.190
be thousands of people. And those thousands of
00:38:18.190 --> 00:38:20.929
people will be trusting what we do. So again,
00:38:21.070 --> 00:38:23.789
coming back to the trust, I would say that this
00:38:23.789 --> 00:38:26.090
is one thing that I believe is going to be harder
00:38:26.090 --> 00:38:28.730
for AI to replace a human. That makes sense.
00:38:28.869 --> 00:38:31.610
And AI doesn't have the ability to build a network
00:38:31.610 --> 00:38:35.190
of trust. Not yet. It's easy to say that tomorrow
00:38:35.190 --> 00:38:37.789
there will be a chat where they will just ask
00:38:37.789 --> 00:38:41.130
answers and they get the offers. I can see it.
00:38:41.130 --> 00:38:43.670
It's coming. It's possible. But I don't see it
00:38:43.670 --> 00:38:47.690
coming that AI is just this network that those
00:38:47.690 --> 00:38:50.780
humans that you're connected. I'm also currently
00:38:50.780 --> 00:38:53.440
building a community where all the borrowers,
00:38:53.440 --> 00:38:55.420
they're discussing with one another their pain
00:38:55.420 --> 00:38:58.619
points and what are the clinic growth problems.
00:38:59.340 --> 00:39:03.599
It's a close community. They find only like -minded
00:39:03.599 --> 00:39:05.719
people. It's not like a Reddit where they just
00:39:05.719 --> 00:39:08.139
post something and someone comes. You don't know
00:39:08.139 --> 00:39:10.059
if it's a legitimate person or not. It's just
00:39:10.059 --> 00:39:12.760
a fake profile whatsoever. So this close community
00:39:12.760 --> 00:39:14.800
where people will be discussing things with one
00:39:14.800 --> 00:39:18.070
another, answering, understanding. This is, again,
00:39:18.170 --> 00:39:21.570
it's hard for me to see AI is replacing it. That
00:39:21.570 --> 00:39:24.969
sounds quite valuable. And is there anything
00:39:24.969 --> 00:39:28.570
that you'd like to talk about? Yeah, what I would
00:39:28.570 --> 00:39:32.469
love to talk about is, especially from the founder's
00:39:32.469 --> 00:39:36.670
perspective, this is a great time to see that
00:39:36.670 --> 00:39:40.880
life, the world is changing. And it's not a choice
00:39:40.880 --> 00:39:44.239
anymore to, no, I don't like AI. And I think
00:39:44.239 --> 00:39:46.559
that it's not an emotional thing. You have to
00:39:46.559 --> 00:39:49.340
do that. There's so many people who say that
00:39:49.340 --> 00:39:53.059
we are boycotting this AI arts is because they
00:39:53.059 --> 00:39:56.179
are stealing the art of human. I agree with them
00:39:56.179 --> 00:39:58.860
that maybe in tomorrow there will be rules and
00:39:58.860 --> 00:40:02.099
regulation behind it. But as for today, it's
00:40:02.099 --> 00:40:05.119
hard to say that I'm boycotting with this and
00:40:05.119 --> 00:40:08.469
that. You have to adapt. So giving you one example,
00:40:08.710 --> 00:40:10.949
I have a friend, she said that whenever I see
00:40:10.949 --> 00:40:14.909
that this video was not produced by a real human,
00:40:15.010 --> 00:40:16.829
there was an AI who was producing this video.
00:40:16.929 --> 00:40:19.510
I don't even watch it. I say that, OK, but I
00:40:19.510 --> 00:40:22.550
say that like it or not, maybe in six months,
00:40:22.630 --> 00:40:24.829
all the videos are going to be produced by AI.
00:40:25.050 --> 00:40:28.010
And so you might the human, he just sent his
00:40:28.010 --> 00:40:31.250
picture and the AI converted the whole video
00:40:31.250 --> 00:40:34.210
with the picture of the presenter. Like it or
00:40:34.210 --> 00:40:37.980
not, it's coming. So my advice is to don't be
00:40:37.980 --> 00:40:43.280
too emotional about AI. This is the evolution
00:40:43.280 --> 00:40:46.639
and we have to adapt accordingly. One thing is
00:40:46.639 --> 00:40:49.639
I agree that always have our values. We should
00:40:49.639 --> 00:40:52.519
not overcome on our values. But I don't think
00:40:52.519 --> 00:40:55.460
there's a value game in someone is producing
00:40:55.460 --> 00:40:58.659
their videos because it's easier. You have to
00:40:58.659 --> 00:41:01.940
go with the flow in other cases. And how can
00:41:01.940 --> 00:41:05.469
we... keep up to date with you and Libora? I'm
00:41:05.469 --> 00:41:08.449
extremely active on LinkedIn, and I started posting
00:41:08.449 --> 00:41:11.630
also on Libora at Instagram. So connect me and
00:41:11.630 --> 00:41:15.550
follow me on LinkedIn. So you can find me with
00:41:15.550 --> 00:41:17.969
my name, Sharmini Akhil, or with my company name,
00:41:18.070 --> 00:41:21.369
L -Y -Y -V -O -R -A, Libora Landing Platform.
00:41:21.670 --> 00:41:27.389
Or you can go to my website, www .lyvora .com.
00:41:29.019 --> 00:41:31.539
So just send me a message, whatever your question
00:41:31.539 --> 00:41:35.420
have about how as a founder I'm using AI or,
00:41:35.539 --> 00:41:38.139
you know, someone who needs to borrow or have
00:41:38.139 --> 00:41:42.679
a loan, I will be happy to help. I started this
00:41:42.679 --> 00:41:45.139
podcast because of an image that came to me in
00:41:45.139 --> 00:41:48.239
a dream. Back when cities had walls, everyone
00:41:48.239 --> 00:41:50.860
passed through the gate. Talk to enough people
00:41:50.860 --> 00:41:54.159
there and you'd understand the city. I wanted
00:41:54.159 --> 00:41:56.340
to stand at the gate and understand not just
00:41:56.340 --> 00:41:58.829
how AI was changing, but how AI was changing
00:41:58.829 --> 00:42:01.849
us. If you know someone building or leading with
00:42:01.849 --> 00:42:05.369
AI, or if this episode was useful, an introduction,
00:42:05.610 --> 00:42:08.409
a share, or a quick review on your podcast app
00:42:08.409 --> 00:42:10.869
all help keep these conversations going and help
00:42:10.869 --> 00:42:14.250
others find them. You can reach me on LinkedIn
00:42:14.250 --> 00:42:17.530
at Daniel Maneri or by email at daniel at maneri
00:42:17.530 --> 00:42:22.030
.house. That's daniel at m -a -n -a -r -y dot
00:42:22.030 --> 00:42:25.150
h -a -u -s. Thanks for listening.