ve-BINAR: Business Development Reimagined

Wonderful. Good morning. Good afternoon. Depending on where in the world you are. Welcome to our next edition of the Vendax webinar titled Business Development reimagined how smart companies are using AI as their fractional growth partner. And we're really happy to have Tarun Thamala with us as a guest who is an expert in that field. A few housekeeping notes. We're going to have a discussion between Tarun and I for the next 25, 30 minutes. Then we'll open it up for Q and A if there are questions. Looking to wrap it up within 45 minutes. So please stay muted while we're in the general session. Improves the sound quality for everybody, but then unmute yourself later on we are recording this. We're also live streaming it to LinkedIn and the recording will become available later today on our website and other social media channels. So let's focus on the topic. Tarun, we've had a few guests in the past six months on the show that had an AI focus. Why don't you introduce yourself and tell us a little bit about how you ended up being an expert on the all matters AI?

Sure, sure. Yeah. Thanks for having me on Henning. My name is Tarun thamal. I'm the CEO at a company called PressW AI. We've been entirely sort of focused over the last two and a half years on this new wave of generative AI and agents and large language mod and really been curious with this concept of how do we help existing businesses transition to utilizing these agents in their normal day to day, having their teams feel empowered, moving faster and really exploring what does that look like? Where is the real ROI in this stuff? In terms of background, I've been in the AI space for nearly a decade now. That's what I studied in school, started my career at a few AI companies and then founded my first AI company actually directly out of out of college. We built that it was an AI product company in like the event management space. We built that for close to four years and then were luckily acquired at the end of 2022. And you know, following our sort of acquisition we just the timing worked itself out where we watched, you know, the birth of ChatGPT in November 30, 2022 and you know, my partners and I were taking a look at the space the world was kind of waking up to what's possible with AI and we knew that we wanted to be involved and helping. We think that there's a tremendous amount, an opportunity to helping these sort of businesses Adopt and implement and integrate.

Yeah. And you're a user of AI, you use it in your own business development. And that's really what we want to talk about and explore and give others an opportunity to learn from it. So I would simply assume that with your first venture you already were using those tools that you ended up developing. But why don't you tell us how your first steps in using AI for your own business development, how those went and what the success rates were.

Yeah, and I actually want to be clear. I think like it's important to note the sort of like history of AI and as it's progressed, like my partners and I, like when we were thinking about AI, you know, three years ago, it looked very different, right? Like we had companies spending, you know, usually companies would spend hundreds of thousands, if not millions of dollars to like create these models and they'd be very, very like specific. Like they would be, we're going to detract, we're going to detect like an address in like this big document and you would spend a bunch of money just training a model to do task. And with the introduction of ChatGPT, the big unlock was this sort of, you have for the first time a somewhat generalized intelligence. And not only is it generalized and OpenAI created that, it was democratized. So it was just through an API and through an interface that they provided, people were able to chat back and forth with it and it was flexible enough to start to be constrained to specific tasks. So when we were thinking about AI before, we would spend a ton of time just building it into our products. And now workflows have changed to where there's a few different levels and I'm happy to kind of talk into that, but just using the off the shelf like ChatGPT. Claude, I have numerous workflows for me, particularly especially on the BD side, doing research on potential prospects all the way through to writing documentation and helping me write what everyone else is already using it for and then making things like that more sophisticated. So doing more prescriptive workflows where having agents look into my email inboxes and automatically draft notes that it get uploaded it to CRMs, like stuff like that is I think where the world is moving more towards. And as I like interact with mature AI companies, or mature in this sense, the more of that they seem to have like, the more augmentation they seem to be experiencing.

We use the word partner growth partner in the title for this webinar and to me that instills that it's not a Replacement, but, but somebody who sits on your side and improves your productivity substantially. Is that how you see it as well?

Yeah, no, absolutely. I think it's just about doing more with less human capital and it's being able to take a single person on your team and make them be able to produce more efficient and not only more efficient, enjoy their lives more. We think about, we're largely a development company, we have a ton of engineers and all of them are using these sort of like AI coding assistants and co pilots and things. And when we do them, not only are they able to produce more work, but their work is more enjoyable. The little nuances of. Oh, how do you write this specific function in this one way? Well, now there's an agent that sits there and can kind of massage away the rough edges behind the workflow, allowing people to stay in the zone and locked in in terms of what the actual big picture thing that they're working on is. And so I think growth partner is a great term for it because if you invest the actual technology itself and like build out what all of these assistants can be and each one is kind of delegated to a certain task, you're still the orchestrator, you're still the, like you still need a conductor. And, but that's, that's what it's shifting more towards is it's like I always, I always bring up the sort of ratio like you know, for the, for a majority of time people spent 80% of their time like generating things and 20% of their time reviewing things. And I think like these agents in this sort of new way flips that right, where you have 20% of your time actually generating things, but an 80% of your time reviewing. And that's like where that orchestration and management layer goes towards, I would say.

Yeah. Now do you. There are a lot of companies who go to market and say, you know, we, we can, you know, we. Instead of hiring 10 BDRs, you hire one and that one does the work of 10, sort of the 10 X factor and companies use different ratios. Is that the way that you think about it as well? Meaning you make you. It's not about increasing productivity. You're not have 10 people doing the work of 100, but you have one person doing the work of 10 and you only pay one or maybe with investment in agents you pay two. So it's about, it's about efficiencies and return on investment. How do you view that for your own business?

Yeah, again like for us, like the easiest one for me and like, the place that I understand being a technologist and programmer for a material majority of my career is on that side of things. And I think it's actually a little bit more interesting than that. I think what we find is that it's not as. It's not as clear as like just 1 to 10x. I think the cool part is that it's actually weighted far more heavily to the people that can use it the best. So if we have a developer that's already a 10x without AI, they become 100x. So it's a multiplier. Right. It's not just a straight ad. Whereas, like, you might have somebody who's maybe more average. Now we have a bunch of 10x people. It's definitely the way that we think about it. It's just the people that get to the people that can use the things better and also just have like, the raw skill to back it up. Like the returns on it are insane. Yeah.

Well, now let's. Let's share the secret sauce of TARUN in business development with our audience. As much as you feel comfortable doing, can you sort of break down for us the. The business development process for your business and where all the different AI agents come into play?

Yeah, definitely. So. So when I think about like sales or business development, it's. It's largely. It's largely information, right? Like, especially when it comes, it's largely information for a few reasons. For one, like, we want to make sure that like, whatever, like, I don't want to sell somebody something that they don't actually want or need. Right. And like, I'd ra the problem where, like, AI and you know, what I'm selling are AI services. So we can help you build like an agent to do xyz. Well, I want to make sure that that agent is actually like, going to solve like a real pain point for somebody and getting like, using them. Something as simple as doing research on a prospect is incredibly helpful. Right? Like just. I can. I have like a series of sort of like, prompts and workflows that everybody that I speak to goes with where I'm looking into kind of what their business is doing. I'm getting a rundown of where they are standing today. Not just a business, but also, you know, as a. In terms of like their AI technology. Like, do they have. They already made investments on it? Are there like press releases out there where they're talking about their AI investments? Are they like, looking to maybe get acquired soon? Like, can I start to like, figure out some of these pieces of information because those become important talking points for me to try or questions even that I can go and ask them about and be like, hey, like, I noticed on your website, like you guys are talking about. Or like in a recent press release, you guys are talking about, like, gearing up for a potential, you know, acquisition or something at some point. Well, great. Like, what, you know, what we should be doing then is like, really looking at your data assets, right? And it allows me to point in a specific way. And this is particularly important for our business because AI is so horizontally applicable, right? Like, you can apply it to every industry or, sorry, every sort of department you can apply. There's so many different problems that you can solve with it. And so for me, it's usually like, we absolutely know that there's going to be some AI agent that's relevant to your business, but what's the right one for you to pick? And the faster I can get to that, the better and the more structured our conversations can be. So from a research perspective, it's huge. And it saves me hours a week, literally, because I just have it set up to be automated to where I get a new calendar invite. It's looking at the attendees, it's looking at the business. It's generating report, it's generating even like a series of. Here are like the, you know, five to ten options that you could kind of present. Here's the script that you can maybe follow and some of the pointed questions you can ask. And then from there, right? Like, where does the next big bucket of my time go towards? It's usually crafting what that proposal or what that initial thing is going to look like. Well, our AI agents understand our business. They understand our case studies. They understand all the work that we have done and can do. And so I can take the literal video transcript. A lot of times I take that, I put it into our sort of like, second agent, the one that's like, helping craft a proposal. And I can go back and forth with it and be like, okay, what types of agents could be relevant for this business? You proposing? Can you help me craft my proposal? It has, like, the proposal template in there. And so a lot of times I can go from a call to a proposal within, like, you know, 24 hours, whereas it might have taken, you know, and that's like me just doing it in the midst of all the other stuff I have going on. It's not a concerted effort. That speed up is, like, significant because I can. I'm not wasting that Much time in between like having the different of these calls, they're not sucking up, you know, 10, 20 hours which a lot of times like you know, BD work is unpaid right until you actually make the sale. And so like all this, all this savings really go towards the bottom line. I can focus on other things during that.

Yeah, I mean that obviously is a lot of content, content coming at you from before the call to point you into the right direction and then content going out. You mentioned the creation and the reviewing share. So you said it shifts from 80% to only 20% reviewing. So that's, that's really what this is about. You still look at it, you still look at the proposal before it goes out. You don't let being sent to a.

Client the like only automations that you know are not, let's call it human in the loop. Like if there's any actual concrete deliverable, whether it's an app or a feature or a proposal or anything like that, a human has to be in there and it has to be, you know, we have to be reviewing it. The whole point though is that when you make that shift from 80% generation, 20% review to the other way around, that entire time gets compressed. Whereas it might have taken you 10 hours to do that 80, 20 split prior now it might only take you an hour and it's like the ratio is flipping and it's still out of a total 100% of time but the total time has shrunk tremendously and that's I would say the really big unlock here. There's definitely a human in the loop. The only times where there's not is where it's automations that are not the most critical or they don't need that much back and forth. So an example would be this research agent, right? Like I don't want to have to go and approve the research strategy every single time there's a new customer call coming up. I'd rather just have the notes delivered to my inbox. That's faster and better for me and I don't need the 99% accuracy in terms of the output. I just need to like get the gist of it so that I can be more effective on my calls versus a proposal. We got to be accurate, we got to be, you know, detailed and that's, that's really like where the difference in like where a human in the loop really enters the picture.

So when you create your prompts or the equivalent of what ChatGPT or other places is called a prompt when you create those. How much tweaking, how much trial and error, how much testing do you do in order to get them as right as you want them to be? Don't need the prep to be 100%, but the proposal has to be. So how much tweaking do you do in order to get it right?

Oh, I think it's constantly evolving, constantly changing. This really maybe alludes to a more fundamental sort of shift that I think people maybe quite haven't understood yet. Some people do better than others. The way that historically humans and we have been buying software for the last 20 years has been in the sense that like you buy software X, like you press a button and then you're getting outcome Y. Like this is not that. This is much more of a management and orchestration type of buy. Right. These are non determined, non deterministic systems and they have to continually like evolve as you evolve. Right. So a better way of like looking at that problem is like, well, if my proposal changes or I want to start including other elements, well then I have to go back to the prompt and edit it. And I want it to be slightly, I want it to be maybe more formal or maybe I want it to be slightly more loose. Like these are all like, it's just management. Right. Like, people don't, there's nobody on my team who I like, don't speak with anymore. Like I just give them a task and now they're done. Like there's, there's constant sort of like iteration cycles and loops and it's the same way, I think, I think treating, treating these AI agents more similarly to just like a human counterpart is like often just, you know, we can put the like more meta conversation aside. Just like from a like, usability standpoint. I think it's the best way of thinking about it is it's like you're continually evolving, but these things grow with you. Right. And they become your assets to help you do your job at least now or in the future better as well.

Yeah. So with that in mind then the question for somebody like myself, who's not in the AI space, who's not a technologist, is always, how much of that do I build myself using a platform, ChatGPT or some other platform, and how much do I outsource to an expert like yourself?

Yeah.

You know, what's your advice to clients, especially in the area of business development in that regard?

Investing in your company and yourself's ability to generate more profit and more money is always a good idea. I think that's like the fundamentals of what we're doing here. And so, but it's not always necessary, right? So sometimes if you have cash in your business and you want to get up to speed as quick as possible, bring in an expert who can help train and that's usually the best place to start. I often even recommend like if a business wanted to simply just be more AI forward and integrate AI more just use the off the shelf stuff, like we don't even need to get complicated and try and build automations and custom software and all that other stuff. All we really need to do is we should be looking at across the board. If you were to score, you know, the company's AI literacy or like use like usage from a scale of like 1 to 10, can we move that 2 to 3 points up? Right. Whatever the average is. And you know the insight I find is like there's a correlation between how effective AI is in a business and how much the executives in the business are using it almost directly to like the more the CEO uses like AI, I can almost promise you, like the more it's used in the business and like their business is much more AI forward. I can find like the types of problems that they're asking us to solve are way more mature and like well thought out. And the reason is because you can get a lot done. Like you can get way better at your current job and like, you know, and I guarantee you just by using ChatGPT and Claude and prompting it really well, like that's maybe 60 to 80% of the lift right in and of itself. And then what you'll find is there's certain automations and this is what we call like level one. Like you like maxing out the usage of these sort of off the shelf tools is sort of what is level one. And then you'll find like, okay, every single time I'm doing the research on this prospect, like in my case, right, Like I'm doing research on this prospect, I don't want to go and copy and paste the same four or five, six prompts back and forth into this thing. I'm going to build a little automation. And that's what we call like level two. And that's just something that's like repeatable, running. But maybe it's like on a no code platform, maybe it's like something like a zapier or something like that. And then, then level three is where we get into the custom software. Okay, Now I want dashboards, I want multiple agents, I want a whole workflow to be orchestrated where my entire process is being automated through a sequence of. Okay, now we're talking a little bit more custom software. Maybe I need authentication and maybe I need observability and true eval. That's where we start to see those kind of uses coming up. The cool part is if you push on that level one, like using the off the shelf tools as much as possible, you'll actually naturally figure out where your like what your level two and level three opportunities are just by like doing it more in the business. Like people in your business will come to you and be like, oh, this like one thing. I have to keep pacing all these problems. Great, now we know what we need to go build. We're not just like firing blindly.

Yeah, well, I, I guess we're somewhere between level one and two as a business ourselves right now and are exploring. Yeah, are, are in deeply, in a deeply exploratory mode. One of the things that you're using yourself are chatbots. I started using Delphi for myself a couple of months ago and trained it on things that are relevant for my business. We had a guest on this show two weeks ago who builds those bots for consultants and coaches who then in.

Turn.

The idea is that the chatbot takes over or takes part, I should say, in that consulting, answering initial entry level type questions. And then the person comes in when the conversation gets deeper and more complicated. Sort of the same principle that you're describing. Can you, can you tell us a little bit more about how you're using chatbots in your business, specifically around business development?

Yeah, absolutely. So, so a majority of my usage is actually like as I've described. I actually don't, we don't really use that many other products like off the shelf product. A lot of our attention is, is much more on the actual development of these tools. Right. So like how do we get more efficient about creating these agents and like how do you decompose a problem into like discrete agents that can go and execute these tasks reliably? That's been like a majority of our focus and interest. Like, to be honest, I haven't. We don't feel a need to like go in and use like these extra chatbots like Delphi and stuff because we've done a good job of like maintaining our like ChatGPT instances and like just building a prompt library that we can use. Plug and play has been like a huge value add for our organization and then anything past that where either we're just building custom for ourselves because we have those capabilities and like we understand the tech and it's easier for us to just go in and make a no code low code automation than it would be to like try and find a provider that's doing something and it's never exactly as we need it. Which I think is actually fairly. Again, kind of goes back to my point of like, I think software has existed in like sort of one paradigm for the last, you know, two decades and now it's kind of evolving into something that is not a like one size fits all. And I think we find this like every single time when we enter a new business. Like it doesn't matter if we built the chatbot that does basically the same, the same thing before connecting to people's integrations, integrating into their workflows in the sense of like getting their actual data in place and like, you know, is this supposed to be an email workflow, A Slack workflow? Is this supposed to be a ui, an interface, a chatbot interface? Like deciding those things is one thing. And even if we do it repeatedly, every single company requires like an extra bit of like finessing and tweaking in order to get it to work to their workflow and to their specifications. Like, you know, think about even things as simple as like, well, I wanted to speak in this voice whenever we're doing this type of thing, or I want, you know, the report that comes out the other end to look like this. Those are all, those all require. Like that's per business. Right. And so it's, it's. Yeah, it's a lot. Like people like you have to like make, you have to train the person to do the job in that business and represent the business correctly.

Let's dream big for a moment. Let's get into that level three because I think if my understanding of your business is correct, that's where you, your business is active. So without names, of course. Tell us a couple of the really interesting problems that you've solved for companies. Things that really sort of brought you to the, to the brink of your capabilities and really challenged you and your team.

Yeah, there's a couple that come to mind, I think. Like either the problems, usually the problems fall into like two categories. One is like it's just complicated to go and get and aggregate and think through all of this data. So this would be like a multi agent problem, let's call it. Or maybe we're dealing with one agent, but the accuracy needs to be insane. And like we're dealing with medical records or we're dealing with, you know, Legal issues where we have to be extremely prescriptive and accurate in our results. And these are obviously, these are AI models, they're not zero or ones, they're somewhere in between in terms of accuracy. So I think two examples of that come to mind. One is actually our own product that we developed. It's our only SaaS. It's our main SaaS product. We do majority of our business through consulting and services work. But we found this one problem in private equity that involves NDAs and a lot of private equity firms, firms see a tremendous volume of them. It's not a problem that's usually worthy of like giving out to external counsel. So people just internally do these NDAs themselves, usually associates and people like that. And so we realized that this was like a fairly prescriptive problem. Like NDAs often have the same type of content. And so we built like a big multi agent system that's reviewing an NDA top to bottom, adding red lines based on like a firm standard playbook. And it's going, you know, it goes paragraph by paragraph, chunk by trunk document or like over across the entire document. It's filling out like where signatures need to be. And so like when you distill this problem down, we have a tremendous number of like these little agents that are running and then checking work of previous agents and that's where we get our accuracy from. And so like that actual problem decomposition is what I call it is like how do you break this workflow down into like little sub agents? Was really interesting. And, and there's another one that's I think really interesting where we're just aggregating data from a few different places. Like we have data coming in from like EDGAR filings, we have data coming in from like medical journals, we have news readings and PR releases. And we're trying to create sort of a cohesive and unified understanding of all of that data as it pertains to specific companies. And that one's really interesting just because like you have have different agents responsible for different tasks. Like some are doing research on the web, some are doing research through documents, some are and one's just handling chat back and forth. These are the types of problems I would say are on the edge where you're really shoving so much data into these models and they need so much data in order to make the correct sort of like next step. That's the challenge is how do you break it down into small enough components and then how do you. Actually I'd say like an even bigger problem is like how do you actually properly evaluate these things? Right. Because it's not like math where there's like a right number and a right answer at the end of it. It's like you're generating an ad copy or you're generating a, you know, a summary of something. Like how do you grade like how good that summary was, like on a scale of like, you know, 0 to 1. And that's really challenging. You need like, you know, a really strong original sort of what we call a golden data set, which is like your proper answers that you're trying to shoot for. And you need to like run these evals and do this is like a huge open area like in our business. I would say.

The example of the NDAs I find fascinating. And you're right, it is. You have to make it very prescribed, especially if you're in a business that signs a ton of NDAs, gets a lot of different versions in. And how do you align all of that sounds like a phenomenal solution to not just the work, reducing the workload, but also making sure it's error proof and you're exactly making your company to something that ends up hurting you down the road.

Yeah, and honestly, I mean I'm going to selfishly promote a little bit, but I think it's genuinely one of my favorite products really from like a, like a meta perspective of I think it just does its job. Like the job of this system is to review a document and get it back out as fast as possible. Like it doesn't like the less number of times that a human needs to like be involved in the decision making, the better. Right. It should. Like this is what I told my team when we first started working on the product. I was like, how close can we get it to one click where we're just dragging and dropping a document in and then it's coming back out. Red line. Like in my mind when I think about like real automations where like intelligence is buried like under a certain, like a certain layer, like there's no chat interface, like you're not interacting with the model at all. Like you're literally just dragging and dropping it in, it's coming back out redlined. You make whatever edits you want and it's going out the door. Like that is in my opinion like one of the cleanest examples of like a true automation that's like wouldn't have been possible three years ago, definitely wouldn't have been possible, but is possible today. And like gets customers value like that and they're not like, like slowed down through random interfaces. And you know, we have like an email feature for example, where you just email it in and it comes back out through email. Like. And you know, obviously I'm close to this problem and it's our own product, so you know, but I just like philosophically, like love the like paradigm that it kind of establishes.

Yeah, well, we're at the 30 minute mark and we're going to open it up for Q and A. But I have one more question and you, your eyes started lighting up when you talked about how you and your team solved this problem. I think AI is too young to have anyone claim that they grew up wanting to build AI models and wanting to be in the AI space. When did you find out or when did you discover that this was the area that you wanted to focus on?

Well, a lot of slight correction. Like AI has been around since like the 1950s in just different forms, right. Like, it's actually been, it's actually one of the oldest like computer science areas. People have been trying to create artificial intelligence for, I think since the beginning of computers. Probably all the way back to like Turing. Right. And that's essentially what he was doing. And you know, we have a famous test now, the Turing Test that tests this stuff. So I, I will make it now the manifestation of AI as it exists today. Definitely. Like, I, this is a newer thing. It's based off of a paper that Google published in 2017 and that's like what birthed this entire era of, of AI as we see it. And so I do think that like a lot I, I like the researchers that I know or the people that I know, like have been thinking about it since they were kids and have been like learning the math and the like hard science that, that goes into AI. I think for me, my fascination started in college for sure. And it started around this concept that like I just kept thinking about the fact that the world is so like not personalized, like impersonal. And that's because like, you know, when you think about an advertising or you think about tv, like you're always thinking about like a one to many, right? Like whenever you see like a billboard, it's a one to many thing. And so I was always, I was just thinking about it from like a man. It would be so awesome if like, you know, I could see the things that I really cared about and like the world was like shaped around whatever I wanted, right? And I think from like a, like technology even. I think like, choose your, like, I've always like been really into like video games. And like, big reason for that is because like the freedom that you kind of have to. And a lot of them to like pick your own path and like choose what you get to work on. And so that personalization, enter the question, like, how do I get, how do you get something like kind of one to. One to somebody, right? And, and make it actually like efficient to do so like in a, in an economic capacity. And that's what AI. Like, that's where I first started getting interested in. I was like, oh, we can, you know, if you take an ad and you run it through all of this math, you can make it actually completely personalized. And in my language that I read and I understand and like, you know, it speaks to me more directly, you know, not to say like advertising is just the easy example. I wasn't thinking about that back then, but, but like personalization of, you know, I want to know like, what are the events for me. I want to know like what concert I should be going to. I want to know all that stuff. And AI is the technology that allowed that to happen. And so that's where it really got interesting to me. And also like just thinking about how the human brain works and neural networks and stuff, I won't delve too much into that. That's where it started. And then once this democratization age came about, like, you know, after the introduction of GPT, I think that's where there's a new sort of fascination of we can build almost a different side of humanity and technology where we have these independent reasoning systems that are able to carry out tasks on behalf of humans and we can make everyone's lives so much better. And people can work less if they want to, they can work more, but no matter what they're doing, their individual output per unit is going up thanks to this technology and we can usher that in.

Great, wonderful. Let's see if we have questions here from the audience. I'm looking at LinkedIn Live so I can pick up comments there. But for those who are live here on the zoom call, just unmute yourself and ask a question. Andrew, you're the first one.

Great, thank you very much. Tarun, thanks. Really fascinating. Congrats on all your achievements and progress. It's quite ironic that while you're talking, just a short while ago, New York Times headline on the administration planning to give AI developers a free hand. So today is is AI Day in Washington. I want to, I have, I have many, many questions, but I'm going to start with just a Couple around information where you know, you were making a couple of points early on about how you know, your sense of sales and business development revolves around information. I don't think it's largely information, but it's one very, very important spoke in a wheel of sales processes. So just focusing on the information piece, I have a sense that as more and more agents are being programmed and developed to, to get information on the other side of the equation, there's, there seems to be less and less freely available good data. So number one is do you, do you actually see that? Do you think that that is correct and that is that would be a continuing trend. And if that is the case, does that imply that for, for companies wanting to get good data on businesses and prospects and ICPs, that we're actually going to have to, to increase our subscription costs to be able to match our agents with good data sources?

That's a good question. That's a great question. Actually it's one that I haven't heard before. So I want to be like super clear about my workflow and I think that the paradigm holds. So there's two things that I wanted to touch upon. One, I do think that more companies are going to understand that their is valuable and they're going to like work hard to protect it. I think that hasn't been like a big debate though. Like it's now reached kind of like a, maybe like a new level because of all of this AI information out there. Like my, my workflow is just for involving like publicly available data, right? Like these are, these are things that the companies want to put out there. Somebody wants to explain what their career and work history is on LinkedIn. A company wants to explain like their new announcement of their new platform, right? These are like public sort of announcements and they, and all it really is like I'd say the sauce behind my information, like you know, my agent that's doing that is not the information that like they go out and retrieve from the Internet. It's our understanding of like how we can best position our services around their business. Right. So like thinking, understanding like oh, this is a law firm, they have these types of workflows. That's a pretty standard like paradigm. The like we know how to deal with legal information based on like our case studies here and like making that association for me ahead of time with, without having me put in my own thoughts on that super valuable. Because now when I get on that call I can be like, oh, tell me about like how your data stored today. Like where are all your cases stored, like where you know, and then I can start to like uncover how an AI agent can impact them. Or they might come to me and be like, hey, we really need your help, like writing briefs or whatever. And I'm just picking a legal example. It would be super helpful for me to know exactly like how we've done it in the past. And like as our projects grow and scale, like our sort of BD team can leverage the same sort of information. And so what you have is like the people that are maybe the best or like have the best understanding or the best information, you can take that, package it and disseminate it to the rest of the team, raising the entire like team's bar up. And that information is mostly just process. Right. And so when I talk about information, it's both. It's you know, whatever's publicly available that's being scraped. But I'm not just reading a report on the company, I'm reading an analysis of how that company could be impacted, impacted by our services. Right. And like how we should position ourselves in this conversation. And that's just like one example. But yeah, absolutely. In the future, I think any company that understands what sort of like proprietary data that they have that somebody else doesn't is going to get, you know, they're going to be very protective about that. But luckily like we have a, we have a lot of like really great protocols like open evidence and things like that that are going to be like open sourced versions of these, of these data sets. And, and my hope is, my prayer is that they'll continue to be popular and they will enable more and more innovation. I don't think protecting and hounding all of this data is necessarily the right move for humanity.

I don't think it's the right move either. But there is a trend, I mean even within new business asset organizations you can see more and more paywalls going up. So then more specifically in terms of top of funnel and to the subject of business development, this is either a level one or a level three question. How, what are your golden nuggets of wisdom on how we can use agents and AI to identify our, our target clients, our ICP? We like, we all have very, very specific ICPs. We know who we want to target. How can we more effectively actually use AI and agents to get to the right people and to, to begin top of funnel conversations and engagement at the highest level and I would just use the example for, you know, outreach, cold outreach, hyper personalized, low volume, you know, so called Hyper personalized outreach. We've tried a few. They're not working very well. Yeah, so what's your, what are your insights there?

Yeah, I mean, okay, so there's, there's two things. One, the off the shelf tools like these AI SDRs or BDRs, they are not. Again, like this goes back to my kind of point that I like to make to people, which is like AI is not a software that you just buy off the shelf and like, you know, enter in a few details and then just like hands up. Great. This is now running. It's. Those things work, but it takes constant effort. Like, like every day. You need to be refining it, you need to be educating it, talking to it. Again, like, same, same with like a person. But they can do a lot of, they can do a lot of, you know, they can be incredibly powerful if you, if you make that investment. We for example, don't. But like in one of those like tools like, like AI SDRs or anything like that, the more important thing for us, because we're a smaller team, we're leaner is just about like where our attention goes towards. Right? And so we, we have an understanding of our icb. Let's take a look at our NDA tool, for example. Well, all private equity firms are not equal. Some are. Some should be weighted higher than others and some should have. And maybe based on its like region or its size or its team members or like the various titles that when I look at a firm I can look at all of those factors and be like, oh, this is going to be a great customer or not. And so most of my attention goes towards, okay, can we build a small agent that just understands our thought process about how we define an ip? And can we rinse and repeat that to an entire list of potential customers? Right. So if you guys. I'm just going to pick an industry like you guys were going after law firms. Well, okay, like what are all of the features, is what I would call them that make a, like a law firm good or bad for you. And can we start to like score each of them? Right? And so like then what I would do is like, okay, guys, on my team, the. We have these 200 firms that we want to go after. These are the top 10. Like if you go and land any like, you know, it becomes really interesting to like have like automated sort of like lead scoring in place. Automated sort of like personalized outbound around on like, okay, this type of company. Like here's the sort of messaging that I would have like Drafting a plan and allowing it to be like, kind of like allowing it to make revisions and updates as the plan goes on. Like those are where I would say, like, there's more lift today. Unless you're just like willing to buy something like one of these AI programs and just like invest a ton of time into it, which has tremendous returns. If you're able to do that.

Gotcha. Great. Thanks to.

Yeah, no problem.

Great. Well to. We're at the 45 minute mark. There are a few more questions, but I'll, I'll. What I'll do is we'll direct them to you. How can people get in touch with you if they want, if they have further questions or especially those who view the recording later and want to connect with, with you?

Definitely. Oh, I mean I'm always eager to just field emails and I'm on. I'm chronically online so feel free to contact me through there with like any questions. Like my major objective right now is just on like the introduction and education of AI to more people. So like the more people like from a. If you're wondering like why. Well, not only do I think this is like the greatest technology ever and I'm super biased, but I think it's like the most exciting thing that's happened and you know, in recent human history in terms of advancement of technology. I think it's also like beneficial to our business for more people to understand how to use this stuff. Like I think like I always, I always joke with my team and I'm like, look, if everyone understood what we do about like what's possible with AI, we'd never have a leads problem or anything like that for the rest of like the rest of time because it's like there's so many applications of this technology and there's, and they're so applicable across the entire organization, even from like an infrastructure component it. So I always like to field questions and talk to people that are interested in it and I also get to learn a lot about their specific industry in their business. And there's a ton of businesses out there that I knew nothing about before starting press. W. So email is for one and I'm happy to kind of link that up. And then we're also, we're actually about to launch like a questionnaire like assessment thing that we're kind of giving away for free that helps people like understand where they might rank in their, in their AI journey and also give them like some suggestions of tools and things that they can start to like, look at and implement within their businesses. Christian on my team who's here is actually the one who's leading that sort of questionnaire. I think from I think we're, we're taking in like 10 people and doing this for free. So if somebody wants to just like take a part of that, we'll provide the link out and yeah, totally for free. There's no strings attached to nothing.

Okay. I assume you're the only tarun thumala on LinkedIn so people can easily find.

I don't know about LinkedIn but I'm sure like I don't know about all of LinkedIn but I doubt that there's many.

Wonderful. Well, thank you for being our guest. Thank you for answering all our questions here. Very interesting session. All the best to you personally and for your business and thank you everybody for listening in. And we're going off air now and look forward to seeing everybody again in a couple of weeks for our next issue. Take care. Bye Bye.