WEBVTT
00:00:00.119 --> 00:00:01.122
Hey, what is up?
00:00:01.122 --> 00:00:04.331
Welcome to this episode of the Wantrepreneur to Entrepreneur podcast.
00:00:04.331 --> 00:00:15.541
As always, I'm your host, Brian Lofermento, and I am so excited about today's guest, an incredible fellow entrepreneur and expert, because I'll tell you what as I was preparing to interview him here today.
00:00:15.541 --> 00:00:31.814
What I really loved about what he wrote in our pre-interview questionnaire is he said I love talking to business leaders with vision who are ready to have us build digital workforce assistance that do real work alongside their human teams.
00:00:31.814 --> 00:00:48.707
Most business leaders don't know how achievable this really is, and I feel like that's such a great teaser to today's conversation in today's episode because today's guest and entrepreneur really believes not only in the future of AI, but in the today, the present, the current of AI.
00:00:48.707 --> 00:00:50.290
So let me introduce you to today's guest.
00:00:50.290 --> 00:00:51.560
His name is Mark Brennan.
00:00:51.902 --> 00:01:02.719
Mark previously has worked at Salesforce all the way back, starting in 2007, which is where he was introduced to a powerful platform, a rapid development environment and super smart people.
00:01:02.719 --> 00:01:07.509
He then embarked on a journey to deliver greatness to struggling IT departments globally.
00:01:07.509 --> 00:01:09.146
He's had some CIO stints.
00:01:09.146 --> 00:01:11.367
He started a few of his own businesses.
00:01:11.367 --> 00:01:24.329
We're definitely gonna talk about Mark, the entrepreneur, as well as the subject matter expert, but his current company, Kynock Labs, is a trusted Salesforce partner specializing in building autonomous agents and related AI solutions for businesses.
00:01:24.329 --> 00:01:36.921
Their services include assessment, gap analysis, business process analysis, an overview of possibilities I love the expansive thinking there Design and build of agent solutions, and ongoing monitoring and refinement.
00:01:36.921 --> 00:01:45.548
There are so many cool things that Mark and his company are up to, and I think that it's really going to paint a picture of how far along we are in the AI race.
00:01:45.548 --> 00:01:50.847
Whether we realize it or not, there's so much that we could be tapping into, so I'm excited about this one.
00:01:50.968 --> 00:01:52.275
I'm not going to say anything else.
00:01:52.275 --> 00:01:55.046
Let's dive straight into my interview with Mark Brennan.
00:01:55.046 --> 00:02:00.710
All right, Mark, I am so very excited that you're here with us today.
00:02:00.710 --> 00:02:01.093
First things first.
00:02:01.093 --> 00:02:01.555
Welcome to the show.
00:02:01.555 --> 00:02:02.560
I'm very excited that you're here with us today.
00:02:04.683 --> 00:02:06.686
First things first, welcome to the show.
00:02:06.686 --> 00:02:10.432
Great to be here, brian, great kickoff and excited to talk to you.
00:02:10.832 --> 00:02:11.114
Heck.
00:02:11.114 --> 00:02:13.996
Yes, the one thing I didn't mention in the intro is that your accent.
00:02:13.996 --> 00:02:18.108
I already know it's going to steal the show here today, so let's dive straight into your background.
00:02:18.108 --> 00:02:19.444
It's a perfect segue, mark.
00:02:19.444 --> 00:02:20.307
Who is Mark?
00:02:20.307 --> 00:02:23.068
Well, how did you get into doing all these cool things?
00:02:29.939 --> 00:02:30.100
Awesome.
00:02:30.100 --> 00:02:31.846
Well, brian, you were asking where's my accent from and all that stuff.
00:02:31.846 --> 00:02:32.829
I don't think about that very often.
00:02:32.829 --> 00:02:42.170
But yeah, irish family grew up in a few different countries and came to America in 1993, the draw of Silicon Valley Just couldn't resist.
00:02:42.170 --> 00:02:50.025
In fact, I left a gorgeous apartment in Switzerland on the shores of Lake Neuchatel to come here.
00:02:50.025 --> 00:02:56.889
My friends thought I was nuts, but I wasn't, because I was attracted to where the cool stuff gets invented.
00:02:56.889 --> 00:03:09.807
I wanted to breathe the same air as Steve Jobs and the likes of, and here we are a lot of, lot of adventure, you know, just to sum it up, brian, I'm I'm a tech industry veteran now.
00:03:09.807 --> 00:03:14.103
I really admire the inventors and the visionaries.
00:03:14.103 --> 00:03:42.656
I have worked for some of the great companies in Silicon Valley and I I found that well, I, you know I was implementing big ERP systems and working alongside big consulting firms and there was a point in the 90s there where I got a little bit disillusioned with big promises, bloated budgets and then disappointment when we didn't really achieve the vision of these big projects.
00:03:43.300 --> 00:03:46.199
And that's why I started to fall in love with mid-market companies.
00:03:46.199 --> 00:04:15.213
And also I discovered Salesforce and that's where everything changed for me, both the speed of delivering value to mid-market companies that are able to make decisions quickly and you're able to talk to leaders who can they have vision themselves and you bring your own vision to the conversation and we create a vision together and then we go do it which I was not finding in, you know, giant companies.
00:04:15.213 --> 00:04:20.067
It's kind of difficult to access those visionary leaders at the very top.
00:04:20.067 --> 00:04:21.550
And so here we are.
00:04:21.550 --> 00:04:43.238
We've been a fantastic journey and kind of collapse is now my third venture, entrepreneurial venture, and we can talk a little bit about the other two, but that's kind of my journey to here yeah, I love that overview, mark, especially because I think it really, right off the bat, showcases your unique vantage point of obviously third entrepreneurial venture.
00:04:43.259 --> 00:04:52.182
You've already talked about your love for mid-market businesses and I want to go into there and the contrast between the big businesses, the S&P 500 that we all see and hear about in the media headlines.
00:04:52.182 --> 00:04:57.305
I would argue that each brings value to the marketplace and to society as a whole.
00:04:57.305 --> 00:05:01.269
Because entrepreneurs, mark, we can move fast with minimal costs.
00:05:01.269 --> 00:05:08.257
We can break things, we can figure things out, we can be the early adopters of technology, of innovations, of approaches, of strategies.
00:05:08.257 --> 00:05:14.432
Mid-market businesses it sounds like they're really those ones that flesh those out and apply that growth fuel.
00:05:14.432 --> 00:05:18.891
Then, of course, the big businesses turn it into a finished product that benefits all of society.
00:05:18.891 --> 00:05:20.583
That's my perspective on it.
00:05:20.583 --> 00:05:22.747
But, mark, you have such an interesting vantage point.
00:05:22.747 --> 00:05:29.980
I'd love to hear you compare and contrast those three different value points and what matters for each of those different business sizes.
00:05:31.483 --> 00:05:32.324
Absolutely Well.
00:05:32.324 --> 00:05:38.535
Firstly, you know we're not here to criticize the titans of industry.
00:05:38.535 --> 00:05:59.170
If I were able to sit next to you know Larry Page or Larry Ellison or Mark Benioff or any one of the giants, and to help them make a vision come true, that would be extremely exciting.
00:05:59.170 --> 00:06:01.382
But that's not who you get access to.
00:06:01.382 --> 00:06:03.226
Normally when you work.
00:06:03.226 --> 00:06:16.384
You work for a fortune 100 or fortune 500 company, especially if they're not in tech, if they're more of a financial services or manufacturing or they're kind of outside of silicon valley.
00:06:17.607 --> 00:06:22.543
The risk is and this is what happened to me is I am not sitting next to that visionary genius.
00:06:22.543 --> 00:06:28.547
I'm sitting with a group of people who are in middle management and they are risk averse.
00:06:28.547 --> 00:06:33.884
They don't like my appetite for new stuff and so I find myself.
00:06:33.884 --> 00:06:35.348
It's not my happy place.
00:06:35.348 --> 00:06:50.690
And, by contrast, when I can connect with a leader, it doesn't have to be the CEO, but it has to be somebody who's got real kind of decision power and is able to make a vision come true In a mid-market company.
00:06:50.690 --> 00:06:57.552
I actually have access to these folks and we can create a vision together and then we can make that dream come true.
00:06:57.552 --> 00:06:58.942
That's where that's my happy place.
00:06:59.644 --> 00:07:00.588
Yeah, I love that.
00:07:00.588 --> 00:07:08.992
Well, I know that one of your other happy places is in the incredibly evolving world of AI and, mark, I want to put AI right in the center of our conversation.
00:07:08.992 --> 00:07:11.307
You heard me tease it at the very top of today's episode.
00:07:11.307 --> 00:07:13.076
All business sizes.
00:07:13.076 --> 00:07:13.297
That's.
00:07:13.297 --> 00:07:17.951
The cool thing about the AI race right now is that we're all trying to figure it out in different capacities.
00:07:17.951 --> 00:07:24.833
Where's that fit in, especially talking to you as a fellow entrepreneur, and then also applying it to the mid-market businesses that we're talking about?
00:07:32.288 --> 00:07:32.870
Where are we today?
00:07:32.870 --> 00:07:33.434
How far along are we?
00:07:33.434 --> 00:07:38.360
That you don't think most people realize it, excellent.
00:07:38.360 --> 00:07:40.684
Well, brian, maybe we need to go back a bit just to see where we you know the you are here, arrow on the map.
00:07:40.704 --> 00:07:49.033
Salesforce came into the AI world about 10 years ago with Einstein, einstein 1, and it was pretty good.
00:07:49.033 --> 00:07:51.935
It was more visionary than really highly effective.
00:07:51.935 --> 00:08:01.863
It didn't really shake the world, but it was a predictive tool and also an insights tool.
00:08:01.863 --> 00:08:09.156
It was able to look at all your data and able to see things that maybe humans don't have the bandwidth or the or the ability to focus, you know, to see those, see those insights.
00:08:09.156 --> 00:08:14.747
Um, and then it was able to, to some extent, kind of connect with some automation.
00:08:14.747 --> 00:08:25.331
Um, salesforce never got into, uh, training a large language model, an llm uh, I think that was genius of them to say.
00:08:27.053 --> 00:08:28.281
That's not where the money is.
00:08:28.281 --> 00:08:41.543
And you know, you and I read the news and all we hear about with AI is really which LLM outperformed everybody else this week and how much money got poured into their compute infrastructure.
00:08:41.543 --> 00:08:44.932
And we wonder how does this affect us?
00:08:44.932 --> 00:08:46.205
How does it affect you and me?
00:08:46.205 --> 00:08:47.705
And the answer is not very much.
00:08:47.705 --> 00:08:54.107
I mean, we can switch from one NLM to another without much friction doesn't really matter.
00:08:54.107 --> 00:09:00.884
If you want to sit in front of a different prompt and ask the same question and compare the two, you could do that all day long.
00:09:00.884 --> 00:09:01.986
It's not very sticky.
00:09:03.720 --> 00:09:07.842
So I think what Salesforce is telling us is that that's not where the money is.
00:09:07.842 --> 00:09:12.801
The money is how are you going to use AI in your business this year?
00:09:12.801 --> 00:09:16.091
And that's where I think the most exciting stuff is happening.
00:09:16.091 --> 00:09:17.416
Think about it.
00:09:17.416 --> 00:09:25.649
The bit about sitting in front of a prompt and asking it to write a document for you or solve a fairly simple problem for you is played out.
00:09:25.649 --> 00:09:30.761
We've been doing that for two years now, and I think anyone who hasn't done that yet is probably living under a rock, right?
00:09:30.761 --> 00:09:38.841
So I think everybody knows how to do that, and it's not like it's not going to get better.
00:09:38.841 --> 00:09:40.626
It is going to continue to improve.
00:09:40.626 --> 00:09:47.461
You're going to be able to ask open AI agents to do a series of tasks for you, not just one.
00:09:48.182 --> 00:10:34.245
But what's really exciting at this moment is, last October, salesforce announced AgentForce, which is essentially agentic AI, an agent that works 24-7 for you, alongside you and your human team, and actually works independently and autonomously, but you're able to monitor it and you're able to keep it safe and you're able to keep it from, you're able to put in guardrails that keep it from hallucinating or saying toxic things to customers, and you're also able to tell it when to hand over to your humans, when it's kind of off its skis on you know it's, the question is too hard or the topic has kind of moved, it's about to move off script.
00:10:34.799 --> 00:10:41.111
No problem, just call, you know, hand it over to a human and it's done most of the work for you.
00:10:41.111 --> 00:10:44.808
So this is really where the revolution is happening.
00:10:44.808 --> 00:11:03.429
This year and I think by this time next year we'll see a large number of kind of leading edge businesses that see the value here and have built a few agents and I've got them working alongside their humans, and we can talk a lot about what this does to the job market.
00:11:03.429 --> 00:11:04.905
What does it do to the economy?
00:11:04.905 --> 00:11:05.576
What does it do to the economy?
00:11:05.576 --> 00:11:08.227
What does it do to, most of all, to the customer experience?
00:11:08.227 --> 00:11:14.477
That's what I think is the most important part of of this revolution that we're in yeah, mark, hearing you talk about that.
00:11:14.739 --> 00:11:16.123
There's two things that I really enjoy.
00:11:16.123 --> 00:11:21.740
One is it's really cool to hear actual use cases of ai instead of just prompting.
00:11:21.740 --> 00:11:27.591
But then the second thing that I want to point out for listeners is you and I we're not talking about years away from this.
00:11:27.591 --> 00:11:33.471
You're saying this time next year, and I think that that just speaks volumes to the rate of acceleration in this marketplace.
00:11:33.471 --> 00:11:44.279
So, with that in mind, mark, when you talk about agents we've all kind of heard that term in a very loose way what are some of those use cases of what these AI agents can actually do inside of businesses?
00:11:44.340 --> 00:11:52.980
use cases of what these AI agents can actually do inside of businesses, absolutely so.
00:11:52.980 --> 00:11:53.842
Let's compare it to over here.
00:11:53.842 --> 00:11:55.787
You've got a person sitting in front of a screen and they've opened an AI.
00:11:55.787 --> 00:12:02.924
They've opened OpenAI or Gemini or any one of the sort of top seven or eight, and they're asking it to do something.
00:12:02.924 --> 00:12:10.336
They're asking it to research something, or to write a document, or to write a poem or a short story or whatever it is that you're asking.
00:12:10.336 --> 00:12:13.830
It can do that for you reasonably well.
00:12:13.830 --> 00:12:29.980
You might need to ask the question a few times until you've prompted it correctly and you've got yourself the right output output.
00:12:29.980 --> 00:12:42.809
An agent, by contrast, is using that LLM, that large language model in the background, but what it's doing is it's tasked mechanically to do certain things to reason with either an employee or a customer and to understand what the intent is.
00:12:42.809 --> 00:12:44.634
What is it being asked to do?
00:12:44.634 --> 00:12:47.840
And to understand what the intent is, what is it being asked to do?
00:12:47.840 --> 00:13:00.114
And then to call on a number of actions that sit inside of its kind of toolkit and call the right one and, at the same time, call the large language model in the background to assemble responses back to you.
00:13:00.114 --> 00:13:02.546
So that may sound really nebulous.
00:13:02.546 --> 00:13:04.471
So let's go straight into a couple of examples.
00:13:04.471 --> 00:13:07.544
The most obvious example is customer service.
00:13:07.544 --> 00:13:21.047
You go to the chatbot on the website for your credit card company or your health insurance company or your you know, pick any number of scenarios and you're asking for help.
00:13:21.047 --> 00:13:22.472
Why are you doing that?
00:13:22.472 --> 00:13:34.071
Well, one, because we're kind of tired of calling call centers in faraway countries and explaining ourselves six times after waiting on the phone and then finding that they weren't able to solve the problem.
00:13:34.071 --> 00:13:37.350
Two, because normal chatbots actually don't work.
00:13:37.350 --> 00:13:45.230
All they do is just point you to the frequently asked questions, and then you have to read reams of documentation and find that you have to solve the problem by yourself.
00:13:45.230 --> 00:14:12.033
So now what we're, the world that we're in, is the chatbot is actually driven by an agent, not a human, and the agent is able to understand what you're asking and go find the information that will resolve your problem if it holds it, and it's using the llm in the background to basically blend private internal information um in in salesforce.
00:14:12.414 --> 00:14:13.378
Let's talk about this for a second.
00:14:13.378 --> 00:14:22.065
Salesforce is probably the only platform in in the enterprise business world that is able to offer this agent force end to end.
00:14:22.065 --> 00:14:23.187
Why is that?
00:14:23.187 --> 00:14:29.703
Because it's a unified platform that's got your customer relationship management data.
00:14:29.703 --> 00:14:40.740
It's got your service data, including knowledge articles about cases, problems, how we resolve them, what the questions are and what the answers to those questions are.
00:14:40.740 --> 00:14:44.206
Nobody else has that integrated in one.
00:14:44.206 --> 00:14:45.730
You know, one pane of glass.
00:14:45.730 --> 00:14:53.991
And now it's got this Atlas reasoning engine, which is able to understand your intent.
00:14:53.991 --> 00:14:58.128
You've gone to that, to the chatbot, or you're texting, or you're on the phone.
00:14:58.128 --> 00:15:09.669
It's able to manage all of these channels and it's able to go fetch internally and then blend with a public AI formulating the answer to you.
00:15:10.071 --> 00:15:13.480
Now the first thing you might wonder is what about your private data?
00:15:13.480 --> 00:15:17.068
Are you training a public AI with your customers' data?
00:15:17.068 --> 00:15:18.131
Well, no, you're not.
00:15:18.131 --> 00:15:25.607
So Salesforce is taking great care to make your data safe and to mask it.
00:15:25.607 --> 00:15:36.804
They've done a whole lot of work on prompt defense and data masking and zero retention so that your data is safe.
00:15:36.804 --> 00:15:50.549
So what's happening here is, essentially, we're able to carry the luggage for the human team and do the more base level tasks and free them up to do more valuable work.
00:15:50.610 --> 00:16:03.048
Now, for example, you may be asking your credit card company about a transaction that took place in Pakistan and you were not in Pakistan this week, so there's a problem.
00:16:03.831 --> 00:16:09.745
And it was able to look it up for you and able to to do so, you know, to find it for you and tell you about it.
00:16:09.784 --> 00:16:14.471
But the next thing might be that might actually be fraud and we might need to hand it off to a human team.
00:16:14.471 --> 00:16:29.157
And the agent is able to is trained, you've told it to do this is to hand it off to a human and to basically prompt somebody in the fraud team, for example, to pick up the conversation seamlessly.
00:16:29.157 --> 00:16:31.874
You don't need to explain yourself again, you don't need to reintroduce yourself, they'll just take up the conversation seamlessly.
00:16:31.874 --> 00:16:32.331
You don't need to explain yourself again, you don't need to reintroduce yourself.
00:16:32.331 --> 00:16:37.450
They'll just take over the conversation and read the summary of what the agent has found so far.
00:16:37.450 --> 00:16:41.168
So let's think about that and unpack it for a second.
00:16:41.168 --> 00:16:42.787
That's kind of revolutionary, isn't it?
00:16:42.787 --> 00:16:55.572
So you've just got half your problem solved by a non-human, by an autonomous agent, and now the human is jumping in and they already know everything they need to know about where we are so far and what to do next.
00:16:56.294 --> 00:16:56.996
Yeah, Mark.
00:16:56.996 --> 00:17:06.592
I love that tangible example because I feel like this is part of the mainstream conversation of people saying, well, yeah, us as consumers, we're going to get solved better, we're going to get solved quicker.
00:17:06.592 --> 00:17:09.586
There's so many big advantages there from a business perspective.
00:17:09.586 --> 00:17:13.794
I'm thinking well, we get to use our human capital in far more intelligent ways.
00:17:13.794 --> 00:17:19.492
We get to take them away from some of those those things that just clog up their to do lists.
00:17:19.492 --> 00:17:24.832
With that said, I also know that the macro argument is what's going to happen to jobs?
00:17:24.832 --> 00:17:26.928
Are they going to shift, Are they going to change?
00:17:26.928 --> 00:17:28.787
The answer, of course, is for sure.
00:17:28.787 --> 00:17:32.625
It is what's your take on what the future looks like in the jobs market?
00:17:32.625 --> 00:17:33.867
Sure?
00:17:34.288 --> 00:17:41.425
well, brian, let's preface this by saying I, I am a techno optimist, uh and uh, but that has worked really well for me.
00:17:41.425 --> 00:18:06.044
Uh, so far, so um, I think what's happening is, you know, you've got to look at job titles and you know income streams that exist today, that didn't exist 10 years ago or 15 years ago Podcasters, influencers, youtubers, gopro influencers.
00:18:06.044 --> 00:18:17.569
Those are just some examples, but there are many, many more include inside of the banking world, inside of the healthcare world, there are job titles that didn't exist before and that do now.
00:18:17.569 --> 00:18:20.839
So let's go back to the automobile industry.
00:18:20.839 --> 00:18:27.189
And how many people in the horse and buggy industry thought they were going to lose their livelihood.
00:18:27.189 --> 00:18:43.615
But what they didn't see is we're not just buying cars, we have car mechanic shops, we have body shops, we have car washes, we have everything related to the car industry.
00:18:43.615 --> 00:18:54.565
The automobile industry has created a massive kind of ballooning economy and essentially you can't argue that there are not more jobs than there were before.
00:18:55.362 --> 00:18:56.914
And this happens time and time again.
00:18:56.914 --> 00:19:02.625
Same with, you know, 1993, 95, the birth of the World Wide Web on top of the internet.
00:19:02.625 --> 00:19:17.596
And you know there were those people who said, oh, it's just a fad, it's not really going to change anything, and that's kind of funny to think back now as the people who thought that is there going to be damage to the job market before there is growth?
00:19:17.596 --> 00:19:18.919
Possibly yes.
00:19:18.919 --> 00:19:28.297
We're seeing already the job market is a little bit tough for software engineers, lawyers and marketers particularly, but also for many other job titles.
00:19:28.297 --> 00:19:29.810
They're kind of struggling at the moment.
00:19:31.246 --> 00:19:36.404
I think the business world has kind of decided hold on the hiring for now, essential hires only.
00:19:36.404 --> 00:19:45.854
We're not sure how this is going to play out and and there are folks kind of seeing oh, they got laid off a few months ago and it's it's not easy to find the next.
00:19:45.854 --> 00:20:20.013
The next thing, I think I heard an amazing line at a Salesforce event last week was tomorrow's jobs belong to today's learners, and so that's suggesting that, you know, give ourselves the flexibility to kind of change lane a little bit and use bringing all of our skills and our experience, our knowledge, just a little bit over here to where we can see momentum, and then we'll be rewarded for that courage and that curiosity yeah, I love that answer, mark.
00:20:20.074 --> 00:20:24.111
Anytime people tell me ai is going to take away jobs, that's the answer that I'm going to share with them right there.
00:20:24.111 --> 00:20:33.593
I think that the car analogy is is so fruitful, especially because a lot of people point towards mechanics, but I love all the auxiliary industries that you just pointed out car washes.
00:20:33.593 --> 00:20:42.786
There's entirely different businesses, business models, services, all of that that pop up because of this, and so it's really cool to think about all the ways that we're going to apply AI.
00:20:42.786 --> 00:21:02.001
When I think about that, mark, one thing that always fascinates me is kind of the yin and the yang, the balance on both sides of things, because when you look at a business, there's areas where we can and should apply AI, but there's also that wisdom and that experienced perspective of knowing where not to apply AI, and I feel like that's kind of where we're at right now is.
00:21:02.021 --> 00:21:04.794
A lot of businesses are asking where do we and where don't we?
00:21:04.794 --> 00:21:11.434
Part of your job and part of your value add with Kynock is that you assess businesses and you look at them and say, hey, let's start here.
00:21:11.434 --> 00:21:16.277
What are you looking at when you look at a business from the outside and when you get on the inside.
00:21:16.277 --> 00:21:21.710
How do you assess where you should be starting and where you should?
00:21:21.730 --> 00:21:22.211
be building AI solutions.
00:21:22.211 --> 00:21:23.073
Yeah, I love that question.
00:21:23.073 --> 00:21:24.195
It's so much fun, you know.
00:21:24.195 --> 00:21:44.421
So we start by saying we're not connecting with as many businesses ready to hit the start button right now and say, let's, do we want to spend X on a few agents that will do these tasks and these and these jobs?
00:21:44.421 --> 00:21:45.366
And we're ready.
00:21:45.366 --> 00:22:02.025
And whether it's experimental, whether it's just a proof of concept type of you know, putting some POC dollars into something that is more of a learning experience or whether we are assured success, we're very confident we can design this, build this and make it work.
00:22:02.929 --> 00:22:06.104
And I think that there's still a lot of I wouldn't even call it hesitation.
00:22:06.104 --> 00:22:13.506
It's kind of still pondering do we have to do this, do what if we don't do it this quarter?
00:22:13.506 --> 00:22:14.667
What if we don't do it this year?
00:22:14.667 --> 00:22:25.875
Our argument is you will eventually be behind and either be left behind or you'll have to play catch up.
00:22:25.875 --> 00:22:32.425
And it's hard to play catch up when the market is ahead of you, when your competitors are ahead of you, and this is the moment right now.
00:22:32.425 --> 00:22:34.810
You're not behind If you're getting started.
00:22:34.871 --> 00:22:38.176
Started now looking at what agents can do for you.
00:22:38.176 --> 00:22:42.232
In start with a corner of your business and start simple.
00:22:42.232 --> 00:22:46.271
We can talk about this some more, like where do you start and how and and?
00:22:46.271 --> 00:22:52.492
How do you assess what would be the the best couple of candidates for agents to get started?
00:22:52.492 --> 00:22:55.185
And then you're not behind.
00:22:55.185 --> 00:23:04.510
Then you know the first agent that you build is a fabulous learning experience for your entire company, and then it will give you the confidence to say, wow, we were able to do that.
00:23:04.510 --> 00:23:11.055
Now, where else can we put agent number two, number three, and you know there isn't a set number of agents that you need to have.
00:23:11.055 --> 00:23:19.413
But you now have yourself a situation where your humans are being assisted by autonomous agents.
00:23:19.413 --> 00:23:20.317
And you've done it.
00:23:20.317 --> 00:23:25.909
You're in the pool right Now you can start to.
00:23:25.969 --> 00:23:36.140
You know, I say this a lot we're limited by our imagination more than our budget, more than our legal constraints or regulatory constraints.
00:23:36.140 --> 00:23:41.289
You know legal constraints or regulatory constraints.
00:23:41.289 --> 00:23:47.482
The real constraint is can you imagine what will you do when you can hire digital workforce agents?
00:23:47.482 --> 00:23:53.565
You can basically extend your workforce digitally, without the headcount and without the payroll increment.
00:23:53.565 --> 00:23:54.887
What will you do with them?
00:23:54.887 --> 00:23:59.676
And so these are, these are great questions, brian, and so I'll give you an example.
00:23:59.696 --> 00:24:00.720
We, we've done.
00:24:00.720 --> 00:24:07.663
We've done a number of fairly basic, either employee support or customer support, the service.
00:24:07.663 --> 00:24:12.066
It's an obvious first place to start if you, you know you've got Salesforce up and running.
00:24:12.066 --> 00:24:15.795
You use cases, right, you have cases for customer support.
00:24:15.795 --> 00:24:28.509
Why not start with solving the simplest and most, shall we say, uninteresting cases for humans, make them interesting for agents and then solve them?
00:24:28.509 --> 00:24:36.294
And it will start off with something really, really simple and prove that it works and then expand from there.
00:24:37.155 --> 00:25:00.345
On the other end of the scale, we've got we're working on one right now with a biosciences research company, little startup, and they're finding that their molecular research is really manual and it's really labor intensive and they're accessing a number of public databases to look up organisms.
00:25:00.345 --> 00:25:11.265
Imagine you start with the end in mind and you want to break down hydrocarbon fuels in filling station sites after they've been shut down.
00:25:11.265 --> 00:25:20.369
You wanna grow a garden where there used to be a gas station, where you can't right now, until you break down those diesel and petrol residues.
00:25:20.369 --> 00:25:29.474
Can I create a natural product that would break down those hydrocarbon residues without chemicals?
00:25:29.474 --> 00:25:31.602
Well, let's find out.
00:25:31.682 --> 00:25:36.756
And so they start researching what kind of organisms have those qualities?
00:25:36.756 --> 00:25:52.683
And weeks later you've done, you know, hundreds of hours of research on these public databases and you've got yourself a list of organisms which, when put together, would interact with each other to form a formulation that would probably work.
00:25:52.683 --> 00:25:55.452
And then you go into lab testing and you actually test it.
00:25:55.452 --> 00:26:00.554
So we're building an agent that will do a lot of that research.
00:26:00.554 --> 00:26:17.431
Well, actually, you start by describing your end goal and it will then go and access those public databases for you and bring back summary of its of its findings, which is incredibly exciting.
00:26:17.431 --> 00:26:22.811
When you think about, this is not just doing a, you know, a low level task.
