Latest / Agentic Conversations (formally mlops.community) / From the Legal Trenches to Tech // Nick Coleman // #332

Transcript

From the Legal Trenches to Tech // Nick Coleman // #332 · Agentic Conversations (formally mlops.community)

0:00
I didn't realize that this was a profession where you helped
0:05
claimants or people that are applying for disability to
0:09
advocate for them through this process.
0:12
And so it was a very interesting introduction because this firm
0:17
handles so many cases. At one time I was literally hand
0:21
handling 300 cases a year. I'm really excited to talk to
0:30
you because of your position, how you are the subject matter
0:35
expert who has turned into a founder that is using AI to
0:43
leverage their product and company.
0:47
And when I say leverage, I mean that you're almost an AI native
0:52
company is what I think the best term would be.
0:55
So let's start with just setting the scene of what exactly are
1:00
you doing, what is the company, and how are you attacking the
1:04
space that you're in? Yeah.
1:05
So in a nutshell, we are providing a suite of tools
1:09
leveraging large language models to help streamline the pain
1:14
points of Social Security disability attorneys.
1:17
Social Security disability attorneys advocate for people
1:22
claimants that are applying for disability benefits 'cause they
1:26
no longer can work. It's a welfare kind of safety
1:30
net in the United States. I'm pretty sure that most
1:32
European countries have these. You essentially pay into a trust
1:36
fund. So if you get hurt, you have a
1:39
safety net. And so if you that that pays you
1:42
a monthly benefit. As with any government process,
1:46
it's a bureaucratic nightmare. And so attorneys are often
1:50
needed to get involved to help advocate for clients.
1:54
And So what we're doing, because of the bureaucratic nightmare
1:58
with documentation, all types of evidence that's needed to help
2:02
advocate for them, we are using large language models to
2:06
streamline a lot of those processes.
2:09
OK. We'll talk about it in the exact
2:10
processes that you're streamlining, but to set the
2:13
scene more for me as someone who is not and has not ever filed
2:20
for any claims. Basically I get injured and I am
2:26
AW2 employee or I I'm a employee or even I'm a business owner.
2:31
I have my own LLC and I can then go to the government and say I
2:35
can't work for the next 6 months or a year.
2:38
I would like to get paid while I'm not working is it?
2:42
Is that what I'm understanding? No, it's generally you need to
2:45
establish that you have a disabled condition that's going
2:48
to prevent you from working for 12 months, consecutive 12 months
2:53
and what they call severe impairments.
2:56
And so it can be a range of conditions from mental to
3:00
physical, whether it's from depressive disorder or to having
3:05
a nerve or compression in your spine, it's going to prevent you
3:08
from working for 12 months. The arduous part of the process
3:13
is getting the evidence to establish why you can't work.
3:18
And that's what attorneys are are doing.
3:21
It's it's it's collecting all the evidence from the different
3:24
people involved in the process from either your doctors or
3:27
getting information from your employers to explain why you no
3:32
longer can either do your pass work, the job that you're
3:35
currently in or even in the past, or why you couldn't do an
3:39
easier job. Walmart greeter, you know, you
3:43
think about like why couldn't you do a very easy, simple job
3:47
and you have to establish it for a 12 month period and you are,
3:51
your payments are based on what you've paid in Social Security
3:54
taxes. So in the United States.
3:56
So when you get withholding out of your paycheck, that's how
4:00
they determine what your monthly benefit would be.
4:03
All right. And now the attorney gets
4:06
involved because if I fill out all the paperwork and then I get
4:11
rejected, that's where the attorney gets involved.
4:13
Or is it when I'm filling out the paperwork?
4:15
There's different attorneys get depending on attorney strategy
4:19
and business model. Well, often times attorneys will
4:22
pick it up initially and other times attorneys don't want to
4:26
get involved till later on because there's a process where
4:30
you have essentially 4 opportunities to establish your
4:33
disability. And then after the fourth shot,
4:36
the agency or the Social Security Administration says,
4:39
hey, you haven't established it. What happens is most people end
4:43
up applying individually online where they fill out an online
4:46
application, provide the medical documentation.
4:49
And then if they are denied, if they call it this first step,
4:53
then they will call an attorney to help them collect the
4:56
necessary evidence and, and kind of navigate this, like, like I
5:00
said, this bureaucratic nightmare, because it's a lot of
5:04
convoluted regulations that don't really make any sense.
5:08
And, and attorneys have to kind of they, they do this all the
5:11
time. So they understand what's the
5:13
most streamlined process to to effectively get you those
5:17
benefits and that we're kind of the translator of, of of this,
5:23
this regulatory framework that. And this is where it gets
5:29
interesting, because that, if I understand correctly, is what
5:31
you used to do. Yes.
5:33
And so I started out my journey. I've been practicing for 12
5:37
years now. When I got out of law school, I
5:40
graduated 2011. It was right after the
5:42
recession. It was a joke.
5:44
You had a better opportunity becoming a barista than you did
5:48
getting a a job at a law firm. And so honestly, at the
5:52
beginning, I was substitute teaching.
5:55
I went one week from reading Cat in the Hat to kindergarteners to
6:00
accepting a job at a Social Security disability practice.
6:04
At the time, I didn't know anything about Social Security
6:06
disability other than my mom had received benefits and that you
6:11
had a Social Security number. I didn't realize that this was a
6:14
profession where you helped claimants or people that are
6:19
applying for disability to advocate for them through this
6:22
process. And so it was a very interesting
6:29
introduction because this firm handles so many cases.
6:32
At one time, I was literally hand handling 300 cases a year.
6:37
Wow, 40,000 miles on the road, going to hearings all over
6:43
Arkansas and the region. And then at the same time, they
6:46
expected me to do what is a more technical part of this job,
6:51
which is appellate drafting. Think about the real deep
6:55
research type drafting, essentially auditing the Social
6:59
Security Administration. So what if a case is denied?
7:03
Then that portion, the job is to identify all of the errors there
7:08
that have been made when the Social Security Administration
7:12
reaches A determination, because at each stage they have to
7:15
provide a notice and explain why you were denied.
7:20
And so I had to do both at the same time, which is a very tough
7:27
ride out of the gate with not knowing anything about it was
7:29
pretty scary and big time imposter syndrome.
7:36
And so obviously with with with being a lawyer, you have to
7:39
worry about malpractice ethics. Can I effectively advocate for
7:47
all these people at the same time?
7:49
And that's where that was kind of the catalyst of how I started
7:53
leveraging technology with this job.
7:56
Yeah. So now take me into the journey
7:59
that you have had into tech. Yeah, so like I said, I was
8:04
scared to death when I got into this practice area, but I've
8:06
always been a computer nerd, always trying to find ways to
8:11
automate processes. And so like I was saying, a lot
8:16
of what this advocacy job is, is developing the record,
8:20
essentially, and getting that evidence needed to establish why
8:25
someone's disabled. And So what is that?
8:27
What is that evidence? That's that's the medical
8:29
evidence. And often times you're dealing
8:33
with thousands of pages of medical evidence and you're
8:36
trying to find a needle in the haystack.
8:40
And so at the time before we had all, you know that the AI that
8:45
we have today, you are digging through thousands of pages of
8:49
medical evidence and trying to dig out insights within that
8:53
evidence because oftentimes there is a regulation that has
8:59
some mapping to a clinical finding, so.
9:04
Wait, so it's not just as easy as me saying hey, my back hurts,
9:10
I broke my back, I can't sit anymore.
9:12
And the doctor signed off on that.
9:14
Not at all. It's sometimes you can have
9:17
three docs treating doctors saying explaining why you're
9:21
disabled, and the agency will still deny you.
9:24
What? Yes.
9:28
So half your job as an advocate is not just trying to find that
9:33
needle in the haystack through thousands of pages of evidence.
9:36
It's honestly developing the record enough so if it is denied
9:43
that you have an opportunity to appeal it to give that person
9:48
another chance. So they could, for instance, if
9:51
a judge denies it. What I the way I'd strategically
9:54
handle these cases, I present every, I try to get everything
9:57
that I can to establish their disabled, but I also have to
10:00
think, what kind of traps can I set in the record that will help
10:07
me get this case appealed and sent back?
10:10
So it's all what you're always looking at it both ways because
10:14
there's only there's, like I said, there's four stages.
10:17
The first stage only about 85% of people get are get denied.
10:22
The second one higher and then they and they acts actually
10:25
after that, you go front of a judge and then there's only a
10:28
40% chance that you get approved at that stage.
10:31
So the to be a really effective advocate, you have to look at it
10:35
real time, say, how can I get them?
10:37
How can I win this case now and how can I set this up
10:42
effectively? So if they do get the nod that I
10:45
can set a record up to get this overturned and give them another
10:50
shot at winning. Wow, well you said something in
10:54
here that just automatically made my a light bulb in my head
10:59
go off, which is finding a needle in a haystack.
11:02
And I'm like, oh wait, they literally had a test for large
11:06
language model contacts windows about that.
11:10
How have you started to incorporate different AI
11:14
processes into this? Because it feels like it's ripe
11:18
for many different areas. 100% And So what we're doing and, and
11:23
in the United States they use what they call IC dent beacon
11:26
codes and I'm sure Europe has a probably similar system of how
11:30
the they bill diagnosis codes, right.
11:36
The Social Security disability framework is built on I CD10
11:40
codes as far as what I was talking about severe
11:43
impairments. And So what we're trying to do
11:47
there are severe impairments and it gets really nuanced then
11:49
there's like severe, severe impairments like OK, and it's
11:53
called the listing of impairment.
11:55
And the whole thing that I was very good at when I was doing
11:58
this work in the weeds, 300 cases a year, was finding that
12:02
needle in the haystack and using essentially Adobe and plug
12:06
insurance to create reositories of words to match the
12:09
regulations. You could almost think of a
12:11
technical sense named entity recognition, I guess.
12:15
And so that's what we were doing.
12:17
And So what I would be able to do is flag words to help me find
12:21
the things that I know I needed to establish to win that case.
12:25
And at the time I said, man, it would be awesome if you could
12:28
have this automated system that did all this because even with
12:31
the plug insurance, it was still an arduous process, but it was
12:35
better than a joke. We all just like a control find
12:39
right to go through a 2000 record.
12:41
It was better than that. And so this is kind of the
12:45
catalyst of of why I created Lex Med.
12:48
And we actually, my one of my best friends that are with the
12:51
law school, we initially came up with this idea.
12:53
And even before Open AI, we were looking at different solutions
12:58
through AWS like medical comprehend and things like that,
13:01
but it still wasn't good enough. And so when Open AI came out in
13:06
November 2022, the things that I was trying to do, I could start
13:13
doing that. And so that's, and I went ham
13:18
essentially using the logic that I've been doing manually and
13:22
creating these mapping techniques and building out the
13:27
logic because of a lot when I talk about logic with Social
13:30
Security, it's very much like a system where you're using a
13:36
sequential process with decision tree, conditional decision
13:41
trees. And so using that and leveraging
13:44
my understanding of that logic and then putting it into which
13:48
things like a mega prompt with large language models.
13:50
And that's how I segwayed into leveraging AI and and starting
13:56
this company. And so it is from the get go, if
14:01
we go back to these different work streams that you had one is
14:05
you're trying to get someone there, you're trying to get
14:09
someone not refused and then you're also appealing refusals.
14:14
And at each step of the way you're building a case and I am
14:19
assuming you're looking at what the doctor has given you and
14:24
you're building the case from various doctors and you're also
14:29
putting into the case these back doors like you were talking
14:33
about for the. Repeal exactly trap since you're
14:37
setting up, yeah, setting up booby traps for appeal.
14:40
And so if then that's what effective attorneys do in this
14:45
practice area is have looking at it both ways to always set it up
14:49
for success in case things go wrong, right?
14:53
And so that's the way I set it up.
14:54
And to throw kind of a rich in this thing when I'm talking
14:57
about these 4 stages of appeals that you go through, the 1st 1st
15:02
2 stages, they're going to deny your case based on what a
15:05
contracted Dr. has said about your client.
15:09
So the government has their own doctors that will review the
15:13
medical evidence as they get it. And then they're obviously very
15:17
conservative because they can't give everybody benefits.
15:19
And so those doctors who have not physically examine your
15:24
client are determining how their impairments impact their ability
15:28
to work. And ultimately those opinions
15:32
become what you have to overcome.
15:35
And so that's like the adverse evidence.
15:38
So they have, you have 4 state government doctors that say you
15:42
can't work. And so it's the attorney's job
15:45
to develop the record, whether it's through treating medical
15:48
providers opinions or diagnostic testing, Mr. is ultrasounds,
15:54
whatever you can, because they to establish it, the framework
15:57
requires both medical opinions and objective findings.
16:02
They do use subjective complaints like why?
16:06
How do you feel? But it has to be, there has to
16:09
be a direct connection of how your allegations are aligned
16:13
with the objective findings in the record.
16:16
So there's a lot of a lot of moving pieces.
16:19
Is this where that mapping and those numbers come in, where you
16:22
say, for example, you have a spine injury and that maps to a
16:28
certain number and the doctors for the state are looking at
16:33
that and saying there's not enough evidence here that that
16:37
number that you're claiming you have is real.
16:40
You hit. The nail on the head and so this
16:42
is what I've been was very effective at over the years.
16:44
They have what they call listing of impairments and one of them
16:48
is that you fit under the musculoskeletal listing.
16:51
So they have a category for each body part.
16:54
And so one thing that I was effective with it was one point
16:57
O 4 regulation degenerative, this disease with nerve root
17:02
compression. And so it has all of these
17:06
qualifying factors that all need to be met within a 12 month
17:11
period. And the assumption is if you
17:13
have all of these factors, there's not a question of
17:16
whether you're going to work. The assumption is, is you're so
17:19
messed up you can't work. But they're very difficult to
17:23
me. And so for instance, and I'll
17:25
use this example of finding that needle in the haystack, one
17:28
point O 4 at the time would require diagnostic imaging.
17:32
So an MRI establishing that you actually have nerve
17:36
recompression in your spot. Then you have to have a physical
17:40
examination that shows you have limited range of motion.
17:44
Then you have sensory, sensory loss, reflex loss, muscle loss.
17:49
And then there's another test called a straight leg raise test
17:52
that indicates nerve compression.
17:54
In a 2000 page record, you might not exhibit all of these
17:57
symptoms in one visit. And so that you had you're
18:01
tasked with piercing essentially a puzzle together to make your
18:06
argument that they meet these requirements.
18:09
And this is where while I was trying to, I could create a
18:12
repository of words, which essentially was control F on
18:18
steroids, right, to actually match those.
18:21
So they're giving me a CHEAT SHEET to see if they met these
18:24
requirements. But with large language models,
18:27
it changed the whole game as far as how I did that, because large
18:33
language models can distinguish between synonyms in the medical
18:37
record. And this is where it's big.
18:39
So where you could use, I guess you could say, regular
18:42
expressions and things like that.
18:44
It, it was very arduous task because medical evidence doesn't
18:48
use the same word to describe one thing.
18:51
And, and your, your question about the back impairment is, is
18:55
a great example. Think about how many times ways
18:58
that you can describe reduce range of motion in someone's
19:01
back limited, reduced. And then there's a, there's a in
19:05
medical, different medical framework, record frameworks use
19:08
different words and there might be 20 ways to describe the same
19:12
thing. And that's where large language
19:15
models really are helpful with that.
19:16
They can distinguish between very similar words that mean the
19:21
same thing under the regulatory framework.
19:26
So first of all, I love your use of the word arduous.
19:30
That's a great vocabulary. Then I'm going to try and
19:32
incorporate into my vocabulary. Then second of all, I think I'm
19:38
getting a better picture here now of you've got these records,
19:41
these medical records that are books.
19:44
It's like Iliad and the Odyssey. You've got all of these
19:48
conditions that need to be met in order for someone to
19:52
successfully pass and get their claims.
19:57
And you are trying to match the two and build up this case that
20:01
says, look, judging on when they went to the doctor in January
20:06
2022 and then they got this MRI and then they did this and you
20:11
show all these things that make it so it is very clear they have
20:18
done everything for whatever it was at one point O 4 that you
20:22
were talking about that. All they jump through the hoops
20:26
or they tick all of the boxes and you are now kicking off that
20:31
type of matching with a large language model parsing through
20:37
all of the records, all of the medical records.
20:40
And then the large language model, I'm assuming is also
20:44
parsing through all of the numbers and all of these codes.
20:50
So whether it's .1 or. Four yeah so you you what we've
20:54
been focusing on and and we're get we're about to build our
20:56
chart vision platform fairly soon as as leveraging what's
21:01
available now as far as using RAG to since they put a database
21:08
of this regulatory framework that's hundreds of pages to help
21:11
match these clinical findings to the ground.
21:16
You know, ground truth is like the regulatory framework and
21:20
then match it with the medical record that changes dynamically
21:23
because people are constantly treating right and things change
21:27
where they might not have met a particular listing in January of
21:32
2025, they now have a new diagnostic test that shows that
21:36
they do have nerve compression. And so that's how we're, that's
21:42
how we're kind of leveraging the large language models with that.
21:46
And I can see a world where you're also, as new legislation
21:50
comes out or as these different tags get updated, that is
21:56
getting updated too. So now you not only need an MRI
22:00
and to have these different criteria being met, but you also
22:03
because they updated something in the database from the state
22:08
side. You need this extra box to be
22:12
ticked and the large language model can parse out that in
22:16
these large bills that come through.
22:18
Again, you're in the nail on the head.
22:20
So what what I said used to with one point O 4A, they made it
22:24
harder in 21. And so now they found out that
22:28
too many people were getting approved under one point O 4A.
22:31
So they've essentially added the requirement that you have to be
22:34
completely compromised in your upper arms so that you
22:38
effectively can't use them or you require a Walker.
22:41
So it's like, OK, you have now they've essentially made it as
22:44
OK, you have degenerative disc disease and you require a
22:48
Walker, OK. And you have you have an
22:50
amputation and you require a Walker.
22:52
I mean, I'm not really exaggerating.
22:54
That's essentially how difficult they've made it.
22:56
So now we've have to find different strategies to get
23:00
around that. But that's what, but that's
23:04
what's so great about technology in these new tools because the
23:08
technology allows us to pivot with the when there are
23:14
fundamental changes in the framework.
23:16
And so we've we're going to be doing that with what our chart
23:21
Vision project and we're already have done it with our now of our
23:24
product that we're doing with our transcription services where
23:29
we're actually doing auditing on hearing testimony where they've
23:34
changed some regulations just recently.
23:37
Talk to me more about that. Yeah.
23:39
So when I like I said this, this Lex Med legal medical, that was
23:46
my this has been a passion project of mine.
23:48
But to get to market early and because of the technical feats
23:54
that we needed to deal with the, the, the security apparatus, the
23:58
amount of money that we need to do to get everything and do it
24:01
right with the medical analysis, we started out and we are right
24:06
now we are transcribing hearings.
24:09
And so there's, I know there's a lot of different transcription
24:12
services out there, but what we're doing is actually adding
24:15
that subject matter expert flair to it and actually helping
24:21
identify and translate words that are often misinterpreted
24:26
with off the shelf solutions by using my context and
24:30
understanding of what the way things should be said and who
24:33
the speakers should be. And so we are using my data that
24:40
I have where we've got actual hearing audio and have the human
24:45
transcripts that act as the ground truth.
24:48
And so we're using that to find patterns in our automatic speech
24:54
recognition model using regular expressions to swap out words.
24:59
And then we're actually adding speaker labeling based on who is
25:03
found in the hearing. And so we released that last
25:06
year and we've taken it to the next step of actually using kind
25:12
of my auditing logic with the way that I review decisions.
25:17
So, you know, we talked about how we're building a case by
25:20
using the medical record to help establish disability.
25:24
This is on the flip side, I'm using my logic to look at how
25:29
the Social Security Administration messes everything
25:32
up to find errors. And we're using large language
25:37
models to actually parse the transcript from the automatic
25:42
speech recognition model and identify all of the errors that
25:47
have happened with expert testimony in the hearing and.
25:51
Now this is to presumably, if I'm understanding it correctly,
25:58
understand the new updated laws or requirements or this is more
26:03
general and it's for any hearing.
26:05
This is the same thing, except a different use case with as as
26:08
far as what kind of evidence we're using.
26:11
And so during an administrative hearing, this is where they got
26:15
the where they've been denied twice the claim.
26:18
It's got to go in front of a judge when they go testify.
26:22
The way they evaluate whether they're disabled is whether they
26:25
can work or not. And so they hire an vocational
26:29
expert who's a expert in jobs to testify whether they can do
26:34
other work. Even if they can't do their past
26:38
job, can they do other work? And they frequently mess this
26:43
up. And so this is where the agency
26:47
uses jobs that don't even exist anymore to find that people can,
26:52
that are that they're not disabled, that there's literally
26:55
jobs that haven't really existed in the national economy and in
27:00
30 years that these vocational experts say they can do.
27:03
So nut sorter, unskilled surveillance system monitor, I
27:09
joke, a Blockbuster rental clerk, you know, it's
27:12
essentially that, right? And so that's not an actual, but
27:16
it might as well be a dinosaur Wrangler, a Model T mechanic,
27:19
you know, just on and on. But there's a regulatory
27:23
framework with this on, on with the vocational evidence.
27:27
And So what we're doing is actually having the large
27:30
language models audit their testimony using this regulatory
27:36
framework and databases like we're using a Sequel light
27:41
database that has information on all these jobs.
27:44
And we're having the, we've just started using MCP server and
27:47
functions to actually reconcile whether the job that this
27:52
experts testified actually exist, whether that person could
27:57
actually form that job, for instance, So you like the back
28:00
impairment that person can't bend at the waist more than
28:05
occasionally. Well, the job that the person
28:08
that the vocational expert said they could perform requires
28:11
frequent stooping. It's something as little as
28:15
that, but attorneys miss out on this because there's so much
28:18
going on in the hearing. What we're doing is kind of
28:21
playing that safety net. And so if it is something did go
28:25
wrong, they can use this CHEAT SHEET essentially to file
28:30
something with the court right after and said, hey, this
28:32
testimony is bogus. It you can't use this to deny
28:36
the case or or if they do deny what you've done is is you've
28:44
stabbed you, you set the record up for appeal so that they can
28:48
effectively appeal that case so they can get another bite at the
28:51
apple if it is denied. And what you're bringing here is
28:56
the speed at which someone can understand when something is
29:01
bogus. Yes, exactly.
29:04
The thing with Social Security disability practices is it's all
29:07
about efficiency. You think of an attorney that
29:10
gets paid on the hourly rate. That's not the way Social
29:14
Security disability practices. It's contingency.
29:17
And so think about going back to me working with 300 different
29:21
clients in the year. You're constantly trying to
29:23
prioritize how you should spend your time.
29:26
All right? It's just the economics of it.
29:28
And so you are often left with, man, do I even put in the work
29:34
to spend on a case that has a 50% chance of getting set back.
29:38
If I, I've got 5 hearings tomorrow, I don't have time for
29:42
this, right? It's that's just the way, that's
29:45
just the reality of the situation.
29:47
What we want to do. And with the whole reason I want
29:51
to build this is to make, I don't want people to have to
29:54
make that decision. I want everybody to, to put,
29:59
tell what every attorney to be able to be that super, super
30:02
advocate so they can effectively advocate for every claimant
30:08
equally. And that's what I, that's what I
30:10
see with AI as being like the great equalizer is that I want
30:16
everybody to be as big of a audit nerd as me and be be able
30:21
to effectively advocate and give their clients the best chance at
30:25
winning. And that's ultimately what in a
30:28
nutshell, what we're trying to do with all these tools.
30:31
It's got to be demoralizing when you have that situation where
30:36
you're thinking how to prioritize and you have to let
30:40
certain ones go. Because even though you know in
30:43
your heart that this person should be getting their claims
30:48
pushed through for the amount of work that it's going to take you
30:53
to get those pushed through, you just don't have the time and the
30:57
probability that it will actually go through it is not
31:02
high. So you have to prioritize
31:03
something else. And that seems like now if
31:07
you're able to transcribe it in the moment and then I'm guessing
31:12
that what you're giving that attorney is like red flags that
31:19
they can automatically look at. Exactly, we're giving them the
31:22
CHEAT SHEET. Here's the deal, most of
31:24
hearing, most attorneys aren't listening to this audio in the
31:27
1st place when it sucks. Nobody likes to listen to their
31:30
own voice like and these hearings you can imagine are
31:32
boring sin. So it's like, so they don't even
31:36
do it in the first place. And So what we're doing is like,
31:39
hey, look, you don't have to listen to these horrible
31:41
hearings. We know you're not doing it
31:42
anyways because when you do, you're just looking for these
31:45
gold Nuggets, these red flag flags, right?
31:48
And since you're not even doing that, you're probably not
31:51
appealing and developed the record anyways, right?
31:53
Because you've got 5 hearings on Friday.
31:55
You don't have time to sit there and scrub and rewind and do all
31:58
this to find these things that you don't even know what to do
32:01
with if you do find them. And So what we're doing is just
32:05
making it so easy that you don't have to listen to the audio.
32:11
They don't have to do the research to to determine whether
32:14
this expert testified that whether, you know, most people
32:17
are going to know a dinosaur regular job doesn't exist
32:19
anymore. But hey, maybe a surveillance
32:21
system monitor that job might exist.
32:24
Let's look, let's have it the system, red flag it and explain
32:27
why it does it. And that's that's kind of the
32:31
brilliance with this. Is that again?
32:34
Yeah, we're giving you that opportunity so you don't have to
32:37
create this mental triage is what it is, right?
32:41
Well, Kathy's got fibromyalgia, you know, she's really hurting.
32:47
But I've got another client that actually has nerve root
32:49
compression and has a Walker. I don't have time to spend.
32:52
I don't have two hours, 3 hours to spend on Kathy's case to
32:56
build the record when I need to focus on Joe's back, right?
33:00
We want to remove that from the equation and that's ultimately
33:05
in a that's what we want to do. And that's that's kind of what
33:09
we are doing with the suite of tools.
33:11
It's not just obviously, it is streamlining them, this boring
33:18
work, digging through medical records, finding that needle in
33:22
the haystack, listening to these fringe worthy audio recordings
33:26
that nobody wants to listen to. But the end goal with this is
33:30
that I want everybody to be that zealous, effective advocate that
33:37
can adequately represent these climates and give them the best
33:40
chance at getting their claims approved because so many people
33:44
are denied for arbitrary reasons, that unfair system,
33:49
broken system. And I don't want it to be
33:52
because the attorney had to prioritize whether they took on
33:56
more a dedicated more time to Joe's messed up back over
34:00
Kathy's, you know, fibromyalgia. So that is powerful.
34:04
What in your mind comes next? Man, no, I just, I wanted to say
34:11
I'm really excited. We are actually going to be
34:15
expanding this this, this tool set.
34:17
We're doing a big fundraise, the investments of strategic law
34:23
firm partners that I'm excited they're going to help scale this
34:26
company. And I'm I'm really excited about
34:30
what we can do in this space. And I've got 2 awesome dedicated
34:34
developers give them a shout out, Phil Cannon and Nick Burka
34:39
and they have been awesome. And the whole product
34:43
development cycle and taking my kind of crazy ideas and actually
34:49
putting them in to the logic to make it work.
34:52
And it's, it's really impressive.
34:53
And I'm very excited and I'm really excited.
34:56
You allowed me to come on today. It's been awesome.
34:59
I think this all is built on the great equalizer that AI can be.
35:02
You know, everybody talks about, you know, the the legal industry
35:05
is very scared of this technology.
35:07
But the way I look at is it's democratizing justice.
35:11
And democratizing justice is a big thing when you're dealing
35:15
with governments that have rigorous, sometimes arbitrary
35:20
regulations with people without the means to be able to
35:23
adequately advocate for themselves.
35:26
And so that's what I think it's very powerful about this
35:29
technology. And if leveraged appropriately
35:32
and ethically, I think that it can be a overall good for
35:37
society.