From Psychology to AI Ft Shorful Islam

In this episode of Visionary Voices, host Aqil Jannaty sits down with Shorful Islam, founder of Be Data Solutions, to explore his remarkable journey from academia to entrepreneurship in the rapidly evolving field of AI and data solutions. Shorful shares insights on the importance of mentorship, taking risks, and the growing role of AI in business. They also discuss the challenges of balancing growth with security in the tech space, alongside the exciting possibilities that lie ahead.

This podcast was edited and produced by ThePod.fm. If you need help managing a B2B podcast, head to https://thepod.fm to see how we can help.

entrepreneurship, AI, business development, data solutions, marketing strategies, team management, risk-taking, mentorship, automation, security

What is Visionary Voices Podcast?

Welcome to "Visionary Voices" the podcast where we dive into the minds of business owners, founders, executives, and everyone in between.

Each episode brings you face-to-face with the leading lights of industry and innovation.

Join us as we uncover the stories behind the success and the lessons learned along the way.

Whether you're climbing the corporate ladder or just starting your business journey, these are the conversations you need to hear - packed with visionary voices and insights.

Let's begin.

Aqil Jannaty: [00:00:00] So welcome to the podcast. Thank you so much for taking the time today. Could you give us a top level overview about what it is that you're currently working on and your journey so far?

Shorful Islam: Yeah, so currently I'm running a company called Solutions. Uh, we're sort of a data ai, uh, sort of company and do sort of software development. We're actually working on quite a few interesting things. As you can imagine, with the sort of, uh, rise of AI and pe everybody wants to sort of integrate it or create products on the back of that.

So we're creating. Things like a, an an AI analyst to help, um, a nonprofit query databases, as well as, um, an AI. Uh, augmented as, uh, assistant for a, um, a surveyor. So basically a surveyor's bought, built a product, uh, or we built a product for him where we'll extract, uh, clauses from leases depending on what the use case is.

So if you are, you know, moving into a property, if you're moving out, or if you need renovations, this. Tool will extract the relevant clauses so you know what the tenant's responsible for and what's the landlord [00:01:00] responsible for. Um, and then we're doing the usual stuff like building data infrastructure and, you know, running reports for clients and building analytical models.

So that's what I, I currently do How I got here actually. Um, my background's in psychology. So, um, I have a PhD in psychology, but I actually specialized in, uh, building predictive models, uh, human, you know, predicting human behavior. So my PhD was on, uh, building a, a predictive model to understand why children have accidents.

So what are the reasons why children have accidents? And if you're a parent, unfortunately, it's. Stuff, the style of supervision you have that leads to, that is the biggest predictor of why children have accidents. But anyway, I, I worked for a few years in the NHS and, uh, governments or public sector, didn't quite find my place.

And then I realized that my. Skill of predicting human behavior was in demand in the commercial space or the private sector. Um, and as you can imagine, you know, with the internet rising around the early two thousands, [00:02:00] there's a lot of behavioral data being collected from, you know, uh, you know, portals and, and websites and stuff.

And so I ended up in companies like ITV in, uh, DDB, Omnicom and uh, Wunderman, where I was utilizing basically. Analytical skills, um, to predict human behavior for marketing, for sales, for product, that kind of stuff. Um, and then about 10 years ago, I started my own company, um, one of, now my third company that I'm doing.

So they've all been sort of, um, UK based with international, um, or offshore hubs and engineering hubs. Okay.

Aqil Jannaty: Amazing. Yeah. No, very, very cool. I mean, lots of different things that we can, we can unpack. I, I mean, I'd love to go back to the start of your entrepreneurial journey, let's say, right? Because I think that transition away from maybe the normal corporate route that people go down into entrepreneurship is, it's a very big leap and it's a very different type of lifestyle that you have.

So what was the, I guess, the initial motivator for you to, to get into entrepreneurship and, and what's the story, you know, behind that transition?

Shorful Islam: So it's inter, I [00:03:00] think it's a combination of many factors. One was obviously, I, I was working at Wunderman, I was managing partner, so I was, you know, on the board there. Um, and at the time my kids were very young and it's very hard to balance one of those corporate lives with having a young family.

Something has to always give, right? You either have to not spend as much time as you want with your children and your family, or you have to, you know, not. Go for all the promotions and I'm quite a driven person. I wouldn't have been satisfied with not, uh, giving my a hundred percent. Now, I, I know entrepreneurship's not an easy ride, but one of the things you had on when I started my own company was that I had the flexibility of how I worked, right?

So I didn't need to do the nine to five. I could work outside of that, that traditional hours where the kids might have been awake and I wanted to spend time with them. And also you could, you could, you controlled the balance, uh, basically of time. So that was one of the reasons that, the other reason was just I, I had built teams.

Um, and I'd done them, especially in agencies. So you had Wunderman, you had Adam and EDDB, uh, [00:04:00] tribal Worldwide, and I was just curious whether I could do it for myself because when you're under a big brand, you sort of think whether it's you who's, who's doing this or whether it's the brand that's just.

Allowing you to be successful. And so fortunately, when I did start my, uh, you know, uh, companies, when I went down the entrepreneur, I was able to build teams and we were profitable. So we'd, we'd grown revenue, um, you know, from zero to, to a couple of million, uh, in a couple of years. So I, I felt that, um, you know, I wanted to prove that to myself.

So I did that. And then se then final thing, actually, I found two good co-founders who I just felt it would be good to work with them, and I don't think we would've worked. In our, 'cause they came from management consulting and I'm from agency world, so I don't think our paths would've overlapped in the traditional corporate world.

So us sort of spinning off and creating our own company meant that we could work together and that was quite nice.

Aqil Jannaty: Yeah, AB absolutely. I think the, uh, the lifestyle factor with entrepreneurship, especially if it's, if it's built the right way, I mean, it's, it's really good, right? You can, you can do what you wanna do in your own, your own time, which is a, a huge difference. [00:05:00] But then obviously the caveat to it is. You need to kind of go out and, and get the business in and then do all those different things as well.

Which is the, which is the tricky part. Right. And so it is an interesting journey. What did those initial few weeks look like for you? Like, you know, when you sign things out?

Shorful Islam: y you know, we, we were quite fortunate is in that, um, we'd started sort of new business activity, uh, before I'd left. And so when we left, we'd sort of. You know, we're, we're on the route to sort of winning a, a really big client, but the first weeks were really much about, you know, building your assets.

So it is, it's, it's amazing like how many presentations you need, make sure the website's correct. Um, you know, and also when you are working with clients, and I'd known this from my agency world and, and my partners from their. Management consulting world, um, is that you need all your documents in place.

So contract, statement of work. 'cause you don't wanna be inventing them as as you go along. So, uh, we spent a lot of time sort of doing that and obviously you sort of learn as you go along the journey, but it was very much, um, very different to what I'm used to. You know, I'm used to sort of managing real large teams and working on [00:06:00] analytic projects.

This is very much about admin, I guess finance, uh, the, the things that no one tells you about when you are starting your company.

Aqil Jannaty: Yeah, it is wearing so many different hats, isn't it? Is you're both sales, marketing and then service delivery and then team management, you know, a manager, all these different things. All in one, which I think is a, is an interesting phase, especially in the, in the early days of, I think start starting a company.

Uh, it can be quite overwhelming, right? When you look at it, it's like you have so many different things to do. Um, so it's really cool to hear obviously that you guys have been successful and you've moved past that point, and then scaled the, scaled the team and the businesses as well. Um, I, I guess as well, during that phase where, you know, you'd made the decision to.

Start your own business and, and go into that side of it. What's maybe like one of those key lessons that you took away from that experience? Because I think for people listening who might be on the edge of, you know, potentially starting a company, um, it would be good for them to understand, you know, what, what are some of the lessons that you'll learn along the way.

Shorful Islam: Yeah, so first lesson, nothing goes according to plan, right? If you think you can have a plan and stick to it, it is not gonna happen. So I [00:07:00] think, you know, you know, we, we thought about doing business plans and all of that stuff, and we decided we just dunno what it's gonna look like. So one of the lessons I learned is that.

You've got to go with wherever the opportunity takes you. Um, and so, you know, in the early days we'd, we did some, um, sort of a business by partnering with other agencies, um, who were sort of, you know, a few years ahead of us or we'd, um, you know, we would sort of, uh, link up with old colleagues and stuff and see if we could go to market together on, on certain propositions.

So you, uh, uh, the lesson is, uh, is learn as you go along. There's no, um. Rule book that you can follow to, to, to end up in success. You've got to pivot. Um, and you've got to understand, you know, where the opportunities might take you. The other thing, and it's a big lesson learned, is that keep on top of your admin and finances.

Okay? Because what you don't wanna wait is until the quarter ends, and then your accountants chasing you for that. And then the a year end, make sure like we were very good. I mean, fortunately one of our, um, uh, uh, [00:08:00] partners was a qualified sort of, um. You know, financial person. So he, he'd worked for Deloitte in, in the accounting and auditing department, so he knew all of the stuff.

And so we had all of that sorted out, so we never had to worry about that kind of stuff. So that was important. And also the other lesson you learned is that obviously there were three founders, so a lot of the early stuff we did ourselves, but when we hired people, we realized we'd need to create some kind of way of forecasting cash flow so that we could pay salaries.

And, and that becomes, even now, that becomes. One of the biggest worries that you have as a founder and and an entrepreneur is when you hire people, you have to worry about salaries Now. Mm.

Aqil Jannaty: A Absolutely. It's, it is so funny 'cause that's quite timely for us. 'cause uh, we're starting to grow the team out, right? And so we've kept a very lean team. There's only three of us. Uh, but we're at a point where we need to start hiring more and everything. And that's the, the one of the projects I'm working on right now internally is how can we start predicting our cashflow a little bit better so we can really plan our, our, our resources a little bit, a little bit more, um, because it is, it is a difficult one right, to try and try and crack, especially when maybe the [00:09:00] market's changing and you gotta obviously improve the business and make sure that's, that's coming in.

But then at the same time, you know, hire for, for that future work that could be coming in as well, is how do you manage that? Do you take that decision now or do you wait until. Maybe it's a little bit too late and you need to hire quickly. Right. It's, there's, there's no right balance to it. And that's the, that's the tricky thing for sure.

And that's something we're going through right now is, uh, is, is how we manage that, that whole process.

Shorful Islam: Yeah. No, I mean it, it is, you need to be brave. I think this is the thing I find that people who. Um, are brave enough to, to sort of go off and, you know, hire people because they want growth. Right? Because you, you, you, you know, it is gonna be a catch 22. If you don't hire, you're not gonna grow. And, you know, and, and if you don't grow, then you just stay where you are.

And you know, you sort of, you know, in a few years time you'd be sort of wondering, well, what was it worth? I could have just stayed in my job and done the thing.

Aqil Jannaty: Absolutely. Absolutely. So, I mean, zooming into, you know, the current day-to-day of, what is it you currently do you know, what does, what does your day look like now when it's running

Shorful Islam: So we have a a, a a hub in Daco in [00:10:00] Bangladesh. So that's where we run most of our sort of data engineer, software engineers, data analysts, data scientists, that kind of stuff. Um, so the mornings usually meetings wisdom, so catching up on projects. Um, as you can imagine being CEO, catching up on finances and budget.

Um, obviously not every day, but, you know, uh, typically in the morning. Uh, but generally it's just sketching up with the team. So with the time zone difference, like wintertime now, or you know, where they're available until 1:00 PM 2:00 PM summertime, they're available until two 3:00 PM So it, it is not too bad.

Um, but the mornings usually meetings with the team and, uh, try and get some client joint client meetings with them. That's quite good. Um, also planning, uh, for new business. So we, just to give you context, so my, my co-founder and I, we've tradition, we've sort of gone out for new business through networks and connections that we have.

But LA last year we decided to just start sort of the new lead type activity and, and go out. So we've, we've got a small team now who, who. Sort of, uh, generating leads through, [00:11:00] um, you know, contacting, uh, prospects like, you know, basically. So, um, so that's new to us. That's, that's kept, you know, something that, that I check in on to make sure the team's doing that.

Um, and then the other thing is we are also like, uh. Doing a lot of calls with clients or prospective clients, um, because, you know, because we don't have the in in face meetings anymore, in person meetings, they're all on on video calls. So it's nice just to check in. Uh, and I think clients also find it easy to check in 15 minutes or prospects.

Anyway. I find it easy to just check in for 15 minutes, uh, rather than having to organize a coffee machine, a coffee meeting, you know, in the past. So, so we're doing a bit of that. We're trying to check in with sort of prospects and clients just to make sure everything's on track and. If there's any new opportunities that we can help them with.

So that's sort of a typical day. I don't actually do any of the work. I, I, I do miss it. I used to build models and stuff, but I don't do any of that anymore. I've got a team who does that, but I do sometimes get involved in the project just to see what they're doing. I.

Aqil Jannaty: Yeah. Yeah, absolutely. [00:12:00] How did you find the transition of, you know, being in the weeds and like building the projects out to then, you know, delegating that to other people? Right, because I think that's something else that's come up through this podcast with other founders is it's very difficult to give control over sometimes of, of some of those areas of work.

And again, like we are going through the same, same transition right now. You know, I've been doing a lot of the service delivery overall, and now we're bringing in people and it's one of those things where you don't want quality to slip, but at the same time, you know, there's gonna be some growing pains with it.

So, so how did you navigate that and, and what was your, your thoughts around it?

Shorful Islam: So, you know, it was hard. I, I don't like giving away control mainly because like, especially if you're an entrepreneur, that's the whole point, right? You are in control, so then you're giving away control of your area of expertise. But one thing I've learned is that a, no one's gonna do it exactly like you, but you've gotta let them do it in their way.

Okay? It doesn't mean that your way's the only way or the right way. They also do it. The other thing is. The, you've just, if you can't trust your team to deliver, then you're never going to be able to run. A company or you know, you are, you're gonna have to be a solo, you know, freelancer or a solo [00:13:00] entrepreneur, you know, solopreneur.

So I think because I always wanted to run a company, I've always managed teams, I've sort of learned to let go. Again, it's not easy, you know, but you learn to let go and you trust your team to deliver. Um, and again, they do do it differently, and that's fine as long as it ends up in the result that you want.

Um, and also the other thing I find hard is that I, you know, I've got certain practices in how I work and. The team don't always mirror that. And I get a bit frustrated sometimes. So for example, after every meeting I like to send an email just to recap the meeting, to say, and not everybody in my team does that.

And I get nervous. I'm like, did you all understand the meeting? Have you got the deadlines? Have you understood what the tasks are? Um, so, you know, but like I said, you sort of, uh, learn to let go, trust the team to deliver. And you know, over the years I've, I've learned that, you know, the, the, the work gets done right, it just gets done differently to how you would've expected it to be done.

Aqil Jannaty: Yeah. Yeah, absolutely. It is. It is one of those, as I say, growing pains, right? Where eventually you get used to it, but it is, it is an interesting, uh, interesting phase for sure. [00:14:00] Um, and so you, you mentioned on the marketing and sales side that you guys are starting to do that. Outbound motion that business development.

So I mean, what type of channels are you working on right now? And the reason why I ask that is because obviously I'm in the marketing space, so it's always interesting to learn what, you know, other people are, are currently doing.

Shorful Islam: Yeah, so you know what, it's, it's interesting. We've, uh, we, we tried last year to do content marketing, so, you know. So we do the conduct that hopefully leads come in, but in, in our, 'cause our work is very much B2B, right? So what we find is that, um, clients come to our website but they don't like, show any interest.

They just come visit and then they go and we, we miss those opportunities. So we have a tool that tracks, um, you know, what kind of companies are visiting our. Uh, and we, you know, send then, so what we're now doing is trying to get the emails for those companies as well as other companies that we think, um, you know, would be interested in our services.

And the outbound is pretty much a a, you know, like I said, my, my background is actually also, I've done a lot of cms, so it's pretty much sort of, uh, outbound marketing, email marketing, uh, with a. A, a, a [00:15:00] mapped out journey of what we expect people to do. So a sequence of email and then follow up. We don't, fortunately for us, you know, we're a consultancy, we don't need that many leads, if that makes sense.

We just need a few. And so we have these targets, which look, you know, from an outside point, they're like really small. But for us, because the ticket size is quite high, we just need to convert one or two a month and, and we're done.

Aqil Jannaty: No, no, absolutely. And and it's interesting as well because what we've learned on the, on the outbound space over the last. Few years is that it has changed a lot. Like what worked about three, four years ago just isn't, isn't quite working anymore. And so especially on the B2B side, it's all about just the relationship angle is how can we get our foot in the door with people who is our, who is our ideal client, and then naturally from those relationships, the, the sales kind of takes care of itself, um, because a lot of people just hate being sold to.

Right. And that's the big thing that we are seeing, um, and that's why outbound is, is it is getting more and more tough. Right? That's what, that's what we are definitely seeing in, in the marketplace. And um, like we've been running campaigns for like Nestle and, you know. Companies like that. So quite big companies and they're all struggling as well to, to really crack it.

So, [00:16:00] um, I, I, you know, I think if you can, if you can get it working and get that channel flowing, then a hundred percent it's gonna work really well. But it's, you know, what's that sequence, what's that type of message that's gonna resonate with the ideal clients? Is, is gonna be the big unlock,

Shorful Islam: Yeah, I mean, as I said, historically we'd had our network approach, so we had people referring us and we'd go via networks, and that is pure relationship building, if you know what I mean. 'cause we're going via a network, a trusted source, I guess. Now, you know, the, the approach is still the same. We're there to build relationships, not to sell to people, but it's getting that foot in the door, right?

Because before the network got you the foot in the door and you had, it was easier. Now it's how do you get the foot in the door? And that's why we, we, like I said, we don't need many. We are very selective in who we go after. Um, and then we, you know, the, the team does a lot of research around the people who are our ideal sort of customer profile.

Aqil Jannaty: Yeah. Yeah, absolutely. Absolutely. So I, I guess in terms of the, the company right now, you know, what's the, the plans over the next few years? 'cause obviously I think a lot of companies are becoming a lot more technically inclined, right? They're looking to adopt a lot of these technologies, especially with AI and all these things.

They don't wanna be left behind. [00:17:00] So I'm assuming on your side then that means, you know, there's some big plans of expansion. So, so what does that look like on a, yeah, on your side.

Shorful Islam: Yeah. So, um, we. So, so there's two things we want to achieve. So one of the reasons we, we created Beata is it's also a social venture. So we wanted to create jobs in Bangladesh. So that's the purpose of having the hub in. Bang. I'm from Bangladesh. My co-founder Keith is not, but he, he bought into the vision and he wants to like, create jobs.

So, you know, part of the reason the Bangladesh office exists is to create, um, jobs in Bangladesh so that people don't leave the country. So keep the brain, the knowledge within the country. Um, so part of our growth is, is to fill that mission. But what we're also finding, like you said, is that. A lot of people want AI stuff and you know, we are ideally suited for that.

We've got a data science machine learning that kind of background. So we are very suited to build AI products, but at the same time, what we're finding is that it helps us as well. So a, a lot of like coding and stuff and solutions are very easy now for our team to get in with. So we are trying to think about how do we.

Structure our team so that they can [00:18:00] utilize those technologies. 'cause what we don't want to be fall behind, we just don't want to be, uh, uh, you know, a a a people type organization providing, you know, resources to company. We wanna provide expertise. Um, and so I think that's an, an interesting one because how do we utilize ai.

For our clients. Right. Um, and have all this guardrails around it so we don't end up in situations where, you know, the machine's hallucinated and we've ended up delivering something. The one area that it's really helped with is, uh, things like, you know, documentation, statement of work, that kind of agreements and stuff.

It's really helped us accelerate that. So before we would have a project manager write the statement of work that's now very much gone faster with sort of gen ai. The other thing we are also doing is we are learning that when we work with clients, we build underlying, um, components that are actually reusable across other, uh, applications.

And so, you know, we've, we're in two minds whether we build that out as a product ourselves or whether we have that as a sort of a toolkit, you know, toolkit of components, AI component that we can [00:19:00] deploy in in other clients. So for example, this. Um, product we built for this surveyor where it extracts clauses, if you imagine underlying components of that is an OCR technology.

Really sophisticated because you can imagine leases were like, you know, they're 50, 60, 70 years old with stamps and been scanned in as an image and all of that kind of stuff. So we are our OCR technology that we've sort of. Uh, curated and built is very good at that. So that is in itself could be a product or we could use it on other client use cases.

So we're thinking of how do we sort of create repeatable, uh, processes and repeatable, uh, products in a product lifecycle. So yeah,

Aqil Jannaty: Yeah. I, I mean, absolutely. I mean, we we're seeing a lot in the, uh, in like the marketing agency space as well is productizing areas of, of the service delivery, because then it becomes repeatable, it becomes more scalable overall as well. Um, and on ai, the, the really cool thing that we are doing is whenever we have some type of internal process or whatever that could be, we could quickly build a, a mini app to, to help us.

Do those things quicker and better as well. Um, and [00:20:00] then it also allows us to serve more clients, you know, with, with, with less resources, let's say overall because we have these reputable actions. So I think productizing parts of the business is, is definitely the move forward because the barrier to entry is so low as well.

You can do things so quickly, um, that it makes sense to do it if, if you, you know, if you can. Um, but it's what those processes are because back when I, you know, started running the business I have now is I wanted to, to really productize and automate as much as I could. But then what I found is that there are elements that you need some type of human in the loop or you need some type of, um, yeah, yeah.

Human to, to take over and to, to help on certain things. So it's figuring out, you know, what, what can you automate and what should you automate as well, I think is the biggest thing. Um, 'cause you wanna go down a rabbit hole of starting to automate this one thing, which takes you actually five minutes a month to do, and it takes you about five days to build it.

So, um, it's one of those, you gotta really balance the, the time, uh, better because the barrier entry is just so low now.

Shorful Islam: And, and also you've gotta determine also how often that use case arises. So sometimes you might just do it once a quarter and it's not worth automating that you can [00:21:00] just quickly do it, you know, manually. But if something's happening every week or every um, sort of day, you know, then it's worth looking at whether it automat.

One things we're actually finding is that. Um, if the process of the ingestion of that process is quite messy, okay, it's very hard then to automate because you, you've gotta spend a lot of time cleaning the data. And what you find is if you clean that data, you don't actually need the automation. It actually works anyway.

Um, so you've just gotta balance out between, you know, just because you can automate it if you should. And secondly. Is, is automation the best approach? Whereas you can, um, you know, fix the core issue and then there there's no need to automate it 'cause it's fast anyway. 'cause you fixed the core issue.

Aqil Jannaty: Yeah. Yeah. No, absolutely. Absolutely. Um, so within your industry as well, you know, what's some of the, the latest kind of AI breakthroughs or, or tools that you're using, which you're really excited about? You know, in the upcoming, you know, few months in, in 2026?

Shorful Islam: Yeah, so I think. The conversational ai. So if you imagine, you know, the combination of gen AI [00:22:00] and sort of voice recognition, those kind of things are making sort of apps much more natural to engage with. So, you know, before you know, you had the Alexas and the um, uh, you know, Googles and stuff, you would have to give a particular command to them.

Now, you know, these apps are very conversational. I mean, we're working with one. Who's making a conversational app on sort of lifestyle food habits and stuff, and it's just naturally saying, this is my family, you know, blah, blah, blah. You know, this is how we are, this is what we eat, and the app understands all of that.

So you're finding those kind of interfaces are much more intuitive and, and more natural, um, for people to talk into. So that's one thing that's, uh, seeing, uh, we're seeing. The other one is obviously the, the agenda or the. Uh, agent AI where you can stack UPIs together, uh, to do, to do a sequence of event.

Again, I'm quite hesitant without putting a person in the, you know, a person in the loop, a human in the loop, only because if one stage fails, then it's just gonna pick up a foul space and carry through. And you know what? It, it, I'm really tempted to automate my, uh, market, you know, my outbound [00:23:00] sort of content marketing because, you know, I, I come up with the idea I can get.

A gen AI to craft it for LinkedIn and stuff, and then I could actually get a Gen AI to, or a, an AI agent to publish it. Right. But I just, I'm a bit hesitant. I need to review it before it goes out, and then if I'm reviewing it, I must just publish it myself. Right, right. I go through the hassle of creating another agent, so.

Aqil Jannaty: Yeah, yeah. No, no, absolutely. It is one of those where that QA element, it, it is really important to do. Um, you know, but a lot of the, the work that we do, we use AI a lot, but again, we, we just have to have that human loop to, to really help. Um, but no, I mean, I think, I think the future's gonna be very interesting in, in 2026 with ai and especially the adoption of ai because everyone I speak to is figuring out, okay, how can they actually add this into their current workflow, even if they've never really thought of.

Building an internal app or building any, anything internal. But now, yeah, people are definitely looking into it more, so I think it's gonna be a very good year on, on your side for sure, because

Shorful Islam: I mean, you're thing AI is gonna be very good at, especially generative ai. You know, the LLLM model is that, you know, historically when people had documents or things that. [00:24:00] Uh, would just be sent to them. They, they, they would get round to it, you know, you'd get round to something. Now you could just have a gen ai, you know, NLM model reading it and alerting you to anything you need to do.

So if you might take your inbox, for example, you know, if you've got a busy inbox like I have, you can either go through one by one and flag what you wanna respond to, or you could just have an AI agent sitting on top of it and just summarizing email and then flagging things that you need to pay attention to, right?

And so, therefore, rather than you. Having to like, you know, uh, assess yourself, this sort of flags it. Yes. Again, you'll go in and you'll look at it. You don't want to automatically send off responses, but you want to go in and look at it. And it, it's, you know, these models are very good at picking up things that you shouldn't be bothered about.

So, you know, when you get messages as generally sales messages, they can summarize them very well and just put them lower down because they don't see anything for you to do in that. Um, and so that would be quite interesting actually, because on the other side, the salespeople, how will they change in order to combat that?

Aqil Jannaty: no, no, exactly. Exactly. 'cause uh, 'cause I use fixer.ai Right. Which is to help the email [00:25:00] management. And, and it's interesting 'cause something I've been thinking a lot about is, you know, the best softwares that I use are softwares where I don't really do anything, you know, to, to, to use it. I'm not logging into a dashboard.

I'm not, I'm not using it. It's just. Doing things for me and like Fixer for example, because I was thinking about this. I was like, I wouldn't even think of canceling it because the value I get is, is massive, right? It sorts me, inbox does it all for me, tells me what I need to respond to. And so that type of value is, is it's just so, so much value, right?

For, for the price that you're paying that you wouldn't even think of canceling it. It's just part of your, your day to day-to-day life now. And so I think, I think the next wave of. AI and, and apps, everything are things that live on the platforms that we already use, right? And so we don't actually see it or, or in terms of input into it, but it'll still be working and we'll still get the value from it.

And so I think that's gonna be interesting to see what other platforms come out of the next, you know, few months and this year. Which focuses more on that. Um, that's, that's also something that we're trying to adopt a little bit more into our workflows, right? And to save our clients having to log into another dashboard, do all these things, why can't it just [00:26:00] live where they currently use and, and, and what they currently use itself.

So

Shorful Islam: I mean, the, the other thing that. I think will happen probably this year or definitely next year, is pre-trained sort of AI models to look out for certain things. So if you imagine, let's say you take Slack, right? You've got your dev team talking about stuff. An AI agent could sit on top of that to flag any risks, any dependencies for the PM or for the.

Project owner, or rather than you having to go through the slack and offer for an update, you can imagine this AI agent could just literally, 'cause it'd be pre-trained to say, look, in a typical project, these are the risks, these are the dependencies that you need to look for. And then as it analyzes that discussion, it could pick those up, right?

It, because that's what it essentially is doing. It's picking up patterns. And you know, normally in, in our conversations, we exhibit pattern that either say the project's gonna be delayed or we are gonna hit a dependency or a blocker. And I, you know, I think there's, I've seen some of these, but I, this.

They've not been as that great. They sort of summarize rather than sort of, uh, give you a, a defined route of a, a plan of [00:27:00] action. And I, I think this year, but definitely next year, there will be these tools that will just make people who sit on top of managing people's lives much easier.

Aqil Jannaty: Yeah, absolutely. What's your, your take on security of these, these AI tools? Because I think that's also another big question people have is, you know, you're giving all this data to ai, which is hosted on someone else's server, not necessarily your own local ones. So, you know, when it comes down to security, you know, how, how far, how far do you, so, so how do you guys work around that and, and what's your outlook on that

Shorful Islam: Yeah, so it's, it is quite interesting because we, um, obviously with the data we work with, we, we have to be very secure. So, for example, this AI agent, this AI analyst that we're building actually doesn't query the data. So you ask it a question, what it does, it formulates the query for a local agent. To then run it on the database, so the data never leads.

So we, there's an architectural way of doing it. There's also other ways of making sure that those models are local models so they don't, uh, go out. Um, so we advise our clients, and again, it's a balance between, you know, speed and security, right? And governance, right. Speed and governance. I think. [00:28:00] You end up in that balance, sometimes it's easy just to get Gemini off the shelf and, and you're up and running with a product quite quickly.

Right. But you have to give away your company information to do that. The, I think the, the, I think when, when things are done by the company, for the company, um, I think there's a lot of governance around that and people do scrutinize it and, and it's done well. The, the challenge is gonna be when individuals in a company are doing it for themselves.

As an employee and they don't realize what the risks are. Because if you imagine, if you post something on social media, you know it's going public, right? But when you po when you give something to a, a, a, an AI agent, you don't know where it's going, so you think it's not going anywhere. Okay. So, you know, if I uploaded, let's say, you know, com contracts, you know.

With salaries of all the employees and just say, gimme a, a benchmark of my employee salary. I don't know who's got that. If it ends up in, you know, Microsoft's world or something, whoever's world, they, they, they sort of have a benchmark of all of your, uh, you know, salary, uh, employee salary. So I think where companies are building four companies, I think they're much more.

[00:29:00] Uh, aware of it and alert to it, and, you know, and I think they're taking sensible steps. I don't think, and many companies saying we're not doing it at all, they just wanna take sensible step. I'm more concerned about the individual in the company who's doing it for their own job role, and the company probably doesn't know about it and they don't know about it because we've not, you know, it's so fast moving.

What are the policies that we, you have to put in place?

Aqil Jannaty: Yeah, yeah, yeah. E exactly that. I mean, how many people are working in corporate have chat CPT open and just adding in documents, which again, as you said is a, is a big security risk, right? For, for the company itself. So it'll be interesting to see maybe what comes up to help, you know, companies navigate that as well.

Probably the, the education side on the consulting, you know, for those businesses probably gonna increase a lot as well. Right? Because they're gonna be looking for those solutions. So, um, I, I think that's a very interesting, uh, yeah, a very interesting side of

Shorful Islam: I mean, I, I, I think, you know, those, um, those sort of platforms that when you join a company, you, you have an educational on cybersecurity, on fraud, on bribery. I think this will be a, a, a mandatory thing. Now it'll be like how to use AI in, in the workplace and manage risk and stuff. [00:30:00] So.

Aqil Jannaty: Absolutely. Changing gears a little bit, I love to learn about, you know, your entrepreneurial journey overall, right? So throughout this, this journey, obviously on, you know, the people around you and everything, it can be quite difficult as well if you're always working, always doing these things. So within your, your life, you know, what are some of the, the key people that have helped influence, um, influence you to, to get the success that you've got right?

Shorful Islam: Yeah, so, so, so there have been a couple of people and I met them actually, uh, most of them when I started ITV. So in that corporate environment, and we've sort of stayed in touch and you sort of need a few type of people. So one type of people you need is someone who just to, um. To tell you what the future might look like, like a mentor, but someone who's been there, who's managed, who's a chairperson of other organizations.

Just to sort of reassure you that the, the, the, you know, the bumps you're going through are, are sort of natural and, and you know, just advise you on how you might go through it. You sort of need that. You also need people who are. You know, for all better words, career coach type people, people who can help you work [00:31:00] through problems.

So you can imagine an entrepreneur journey. There are things that you know, and there are things that you definitely don't know about. And when you hit those, what do you do? Right? And the problem is, maybe no one else can help you, but if you have someone who you can talk it through, who can help you, help yourself, then I think that's much better than just going and getting a next.

Expert to solve it for you. Because a, I find whenever you get an expert, you spend as much time with them to try to brief them anyway. So it's better to actually have someone, you know, who, who can coach you through that. Uh, and sometimes that could be like someone who's a close friend and, and I've got a close friend who, who can do that.

But sometimes you might find just, you know, pay for coaching, pay for those kind of career coaching, um, because it will help you. Um, and then you actually need, interestingly, a network of peers. And the reason I say that is because. Uh, you need people who need to know the now, right? So when you come across problems or you come across challenges, you need to hear from people who are experiencing at the same stage that you are experiencing it.

Not that they experienced it five years ago, because the world is very different, right? Five and five years ago, and you don't need junior people who not who wouldn't [00:32:00] appreciate where you are. So I think you sort of need that. Peer network and I've managed to keep in touch with a, a few people that I've worked with, um, you know, so who are my peers and we meet up for quarterly lunches and dinners and stuff.

And it's just nice to have that network of people to ensure that, you know, because some of 'em are still in corporate, you see. And so just to gimme that sanity check that what I'm doing is I'm not falling behind or it's not different to what they're experiencing in a larger organization.

Aqil Jannaty: Yeah, absolutely. And I think as well as an entrepreneur, it can be quite lonely as well, right? Obviously when you have co-founders it helps. At the same time, it is still quite a lonely journey and so I think, as you said, having that peer network of people, maybe at the same level, maybe a little bit further ahead, but at the same time you can have relevant conversations and, and keep each other in check.

I think for me that's been, you know, one of the biggest things. 'cause I remember when I started my very first, uh, market agency, I just thought to myself. If I can make a couple grand a month, I'll be, I'll be chilling, you know, I can, I can just have all my time and, and do whatever I wanna do. But then it wasn't until I, you know, got introduced to these people doing, you know, [00:33:00] quite big numbers and I was like, wow, that's, that's possible.

And then, no, we're friends today, right? And, and we help each other out. And we, we're now all at the same level, which is, which is really cool to see. And so I think having that pig group, especially as an entrepr. It's just so, so important. Um, you know, so you can just stay in touch with those people and, and, you know, understand what the market's doing and all these different things.

But that all comes from the network that you have around you. So you can't, you can't go to go at, at it alone. You have to have those, uh, those people around you for sure.

Shorful Islam: And, and, and it's interesting because that, that network for me, for example, is both other entrepreneurs, but also people already, still in corporate job. But you've gotta build that over a longer period. You can't sort of suddenly go, I'm gonna be an entrepreneur today and have my network. You've gotta build your network first, right?

And then, then go on the journey.

Aqil Jannaty: Yeah, a hundred percent. So, um, one of the, uh, one of the final questions we always ask guests in the show is, if you can go back to your 18-year-old self and only take three lessons with you, whether it's, um, philosophical knowledge, some business knowledge, some general advice, what would those three things be and why would it be those things?

Shorful Islam: So the first one I would take is take a risk. Okay? [00:34:00] So when you're from a working class background, you know, and your, your education system's very structured for you to get a job, risk taking is quite hard, okay? You, you, you, you don't think about taking risks because that's not something that, that you are taught to do.

So to my 18-year-old self. Definitely take risks if they come up. You know, what have you got to lose? Right. Especially when you're young. Okay. That's one thing. The, the second thing is find a mentor early. I mean, I, I, I did a PhD, so I was in academia for a while. Um, and one thing I, I lacked. Is I think a commercial or a, a business mentor.

And the reason I say that is because I, I went for a traditional career route and I had ideas, you know, five, well, 10 years before I started my entrepreneur journey that I should have actually spun off as products. I look back now and I think I had great ideas, but because I was in a company, I just felt that's what, you know, if the company doesn't approve it, you don't, you just move on with your job.

So if, but if I had a mentor who could have told me, that's something you could actually spin off at a side hustle or something because you, you don't know what you don't know, right? So that's why mentors are there. [00:35:00] To do that. So find yourself a mentor. So if you're 18 years old, reach out. You know, the interesting is I do a lot of, um, uh, mentoring myself, and I say to everyone, especially on my YouTube channel, I say to them, look, just reach out to people.

If 19, if 20, if you reach out to 20 people and 19 say no, and one say yes, that's great. You've got, at least you've got, because you need one mentor, right? So, you know, just get that one coffee, even if it means 20 or a hundred meetings, just do it. So get a mentor. And then the final thing, um, I guess I would say is. Actually, I would say, um. Learn something that you don't think you're good at. Okay. Um, so I, I never, I mean, it's ironic I'm in data, but I never actually thought I was good in maths, ironically. Um, I, I, you know, and when I did my psychology degree, I realized there's so much statistics involved that I eventually became good at it, or I, I probably realized I was good at it.

By being forced to do it. But if you are 18 and there are things that you think, like a lot of people will think, I'm not good at coding. Right? Do it, see what happens. You know, they might think that they're not good at marketing. Do it [00:36:00] and see what happened. Especially sales, I mean, I'll tell you now, I never thought I'd be a good sales per, or I'd never thought I could do sales.

But when you're pushed out there, especially when I was at Wonderman and I had to. Do sell stuff for the agency. You realize, I, I actually quite enjoyed it. I love going out, talking to people, you know, selling them the vision, that kind of stuff. So, you know, so learn, you know, go for that one skill or learn, try and learn that one skill that you think you're not good at, or one or two, and you never know, you might enjoy it.

Aqil Jannaty: Amazing, amazing. Marla, thank you so much for taking the time today. I've really enjoyed the conversation. Lots of golden nuggets for everyone listening. So thank you so much for, for taking the time.

Shorful Islam: No worries. So thanks for having me.