host intro

Hey, welcome back to the show. This is John Steele. This is the Funnel Vision Podcast, episode two. And we got a good one for you today. Our guest is going to be Jordan Blackwell. He is the co-founder and CEO of Sia Labs. Sia Labs was founded right here in Philadelphia and they focus on building AI systems, AI-driven systems for small businesses. But their goal is a little bit different than a lot of the other agencies floating around.

are really focused on helping founders remove themselves from businesses, either by preparing them to be sold or by helping founders build systems that can work without them, things like that. And they've talked a lot about...

you know, helping companies build succession plans and be passed down to the next generation, things like that. And so I want to talk to Jordan about that. He also co-founded Sia Labs with his wife and I would love to dig into, you know,

What that dynamic is like working alongside your spouse and sort of founding a new company. They have a really interesting story too as to how this company was founded and and why they have that focus. And they're bringing some of those stories to Philly Tech Week. So Philly Tech Week is a big tech event here in Philadelphia that's coming up in early May and you know Jordan has launched this really interesting event that is focused on how AI is going to help out.

into the future and why Philadelphia is in a unique position to be a part of that change. So I'm really excited to talk to him about that. The event is at capacity already. I believe they sold it out. So good for him on that. But I definitely want to dig into some of the interesting points that he raises in the literature about that event. So very excited to talk to him and hope you enjoy this conversation as well.

interview

Jordan Blackwell, welcome to the show.

Jordan Blackwell (01:59) John Steele, thanks for having me, man. Thanks for having me. Good to be here.

John Steele (02:02) Absolutely. So I wanted to talk to you today about Siah Labs. ⁓ You know, with full candor, I feel like we are in similar areas. Like I feel like we're, you know, we're both kind of talking about systems and helping founders sort of organize their businesses and move away from, you know, being the bottleneck as you say, but you have a really interesting kind of take on it. And I really want to, you know, dig into that a little bit and also.

talk about the Philly Tech Week event, because I think it's going to be a great one. It's just a really fascinating topic. So definitely had some questions about that, but I wanted to start with a real easy one. So can you just tell us a little bit about what Siah Labs does and how you came to found it?

Jordan Blackwell (02:45) Yeah, no, for sure. For sure. It's funny. I tell my wife, I don't even know if that's the easy question because we're always, you know, tweaking the copy just to get to the core of what we're doing. But essentially we're an AI native business operations firm. So what we do is we serve business owners that are stuck kind of running everything themselves and then we use AI to capture what they know to centralize data, handle routine tasks and work so that the owners can achieve.

ease and so our little framework is expansion, acquisition, succession, and or enjoyment. so yeah, started that in 2024. There was, you know, no AI a part of it back then, but it essentially was going and working with a small to mid-sized firms to automate those processes to having a centralized portal to where it's kind of a one-stop shop for all of their internal business operation.

operational needs. And so it could be anything from running commissions to managing their and hosting their websites to having client portals for their customers to go in and access information securely. Yeah. It really was. Let's take a look at all of the bottlenecks within the business and utilize software and now even more so AI to remove those bottlenecks. So yeah, started at 24.

In July of 25, my wife came on board and so now we're co-founders and it's been, ⁓ it's been a blast. It's been a blast.

John Steele (04:20) Yeah, as you mentioned, your wife came on board and I understand that, you know, she, her family was, was in the oil business. You know, can you, can you talk a little bit about that? Cause I'm, wondering if that, you know, how that sort of plays into your, you know, a lot of what you talk about is, business owners sort of building succession plans or, you know, family businesses, things like that. And, ⁓ did that sort of start with her, her family business in a way?

Jordan Blackwell (04:47) So it didn't start with it, but it's very much inspired by it. So her grandparents, Nate and Gloria Hartley, she's from Portland, Oregon. And so he had started back in the 70s. Yeah. The 70s around like 1970s started an oil company. essentially providing residential, commercial, and then industrial oil contracts to homes and, you know, city and state, you know, and even.

you know, national federal contracts. And so yeah, over the next 40 to 50 years, he was just a mainstay in the community and he kept working, you know, that job until he was unable. Unfortunately, he passed earlier this year. And so with that story, it was one of those things where they had an extremely successful company, but it was also one where, you know,

And it was owner dependent where if they stepped away, the business stalled. ⁓ so Siah Labs really wasn't a thing, obviously during their tenure, but it's one of those things that we always keep in mind as we're looking to help other businesses where we want to ensure that no matter what happens, the business can be self-sustaining. if...

something happens either intentionally or something under the blue happens. These businesses are prepared to continue running, whether it be to the next generation or to a third party. yeah, Hartley Oil, a neat Hartley Oil and company is definitely one of those stories that is continually, you know, in our minds as we're thinking about Siah Labs.

John Steele (06:29) Yeah, definitely. Yeah, I found that to be ⁓ an interesting thing, ⁓ especially given your things that you focus on and talk about on your site and to this event. I wanted to ask you about working with your wife Sue. So you said Deanna came on in 2025. How did that sort of come about and what's it been like working with your spouse, co-founding a company with your spouse?

Jordan Blackwell (06:39) Absolutely.

Yeah, no, great question. And it's funny. It seems to be one of the most polarizing, you know, parts of the company where they're like, man, you're, you're working with your wife or you're working with your husband. But there, there are a few things. One, we are, let's see, there's so many things that I want to talk about. So first we're going on nine years of marriage. And so, you know, we've definitely had a ton of time to learn each other, to learn the quirks, you know, each other's strengths and whatnot. And so we, definitely have some.

time on the books to where this isn't, you know, anything new for us in terms of spending time together. and then we just told this story earlier today where Siah Labs is actually not the first company that we went in on together. When we were first married, we started a bow tie business and we grew that all the way to a sale of one bow tie in that 10 year. But it was a, it was a fun journey just, you know.

learning about how to build something with one another. And so kind of the seeds of entrepreneurship were always in our, in our minds and a part of our relationship from the beginning. so, you know, long story short, it's been excellent. It's been one of those things where it's a night and day difference where before it was either I'm servicing current customers or I'm thinking strategically about the business. But now, you know, with two people working full time on this thing, I know that, you know, if

I'm coding some new feature or having a conversation with you. There's the other half where, you know, she's thinking about the business. She's, you know, reaching out to this person or that person and it just makes all the difference. So I really do think of the business as kind of like a ⁓ pre-Diana and a post-Diana. Cause having that extra mind, you know, it makes all the difference.

John Steele (08:45) Definitely, definitely. ⁓ yeah, you, you touched on it a little bit earlier, but I wanted to get into it a little bit more. you, Siah Labs is focused on building systems and you have a framework that you work from called Ease that I was checking out. So, and what does Ease stand for and why was it important to you to highlight things like succession and enjoyment in your, core of what you guys do? I have

I've reviewed a lot of different agency frameworks, you know, in, in researching for, John Steele Consulting. And I haven't seen many people focus on either of those two things before. So I was interested to hear, you know, chose those.

Jordan Blackwell (09:24) No, that's a great question. It's a great question. Ease is, know, expansion, acquisition, succession, enjoyment. ⁓ and so it was a combination of a few things. So first was we simply looked at all of our current clients, because it really wasn't until, ⁓ early 2026. So just a few months ago when we made the concerted effort, like, Hey, we actually need to market, you know, it can't be enough to...

go off of word of mouth and referrals is like, need to get out there. need to shake hands and whatnot and tell people what we're about. And so that led us to, so if we're marketing, like what are we marketing? And so we were taking a look at our current customers and we've had a couple situations where it was a business that was recently acquired. And so the original owners had left, there's a new operator in place and now

They need to figure out, you know, how to systematize what was going on. A lot of the stuff was kind of handshake deals, not really written down, documented. so a lot of our work was, you know, doing that documentation, ensuring that there is no kind of tacit knowledge that's stuck in somebody's head where everything is available for, you know, the people that need that information. And so that was kind of the, the acquisition side, because we noticed, okay, so it was, the acquisition was.

John Steele (10:42) Yeah.

Jordan Blackwell (10:48) the trigger for them needing a solution and to kind of professionalize the business. And then there was another case study where it was a current customer where no acquisition, but it was more right. I'm kind of the owner saying that he's kind of the bottleneck for the entire business. So you can't really scale, you can't really grow because if you add two more people then

All the systems kind of shut down because you're running it off of spreadsheets. And so we're like, okay, so somebody who's looking to expand the business to grow the business. And then what was it? It was actually a friend of mine because the succession piece is the only facet of the ease framework that we haven't had a customer, you know, kind of looking to go to the next generation pass down the business. But we learned about two things. One, entrepreneurship through acquisition.

which is this whole ecosystem of, you know, kind of like recent business grads. Sometimes it's, know, kind of ex private equity, ⁓ folks that are deciding to get into business ownership, but they don't want to start from scratch. They actually want to acquire businesses. And so you have that. And then you also have what we're talking about this upcoming week, which is the great wealth transfer, the silver tsunami people call it where, you know, the, older generations, the baby boomers, the side generation, they're getting older.

And those business owners are starting to think, you know, strategically and seriously about the next step and the next stage of the business. And so we just kind of see this wave coming and we're like, okay, it seems to match a lot of what we're already, you know, doing for our current clients. then enjoyment is kind of, it kind of pulls all of that together where if you are the bottleneck, like what I say is that you've already done the hard part. You've already built a business.

And that's what all of us are trying to do. We're trying to build something, you know, uh, that lasts. so these business owners, they've done that. Now it's just that final step of getting yourself out of, you know, the, the day to day. So you can just take a step back, whether that's thinking more strategically about the business or just allowing the business to run where you, you know, take a step back and enjoy, uh, the fruits of your labor. so, yeah.

It was kind of a culmination of all those different pieces coming together.

John Steele (13:11) Gotcha. Yeah, I should mention, know, like I said before, we're sort of in this similar area. And I know you must have noticed this too, is ⁓ a lot of businesses, owners, consultants, agencies, we're seeing a ton of new operators flood into our area of the market. And I'm just wondering if you worry at all about like an AI bubble, if you think, you know, it's getting too, you know, saturated or anything like that or...

how you sort of think about differentiating yourself. think having a framework, you know, can be a great way for that. And I think a lot of consultants and agencies know that well, is if you have like a way of looking at things, it's a little bit different. And it sounds like you guys have obviously been really thoughtful in thinking about the folks you serve. But what are some other like differentiators that you see and are you concerned at all about like bubble here or just it getting over saturated people kind of looking at it like a fad or something like that? mean,

I've heard people say that about agencies in this space and definitely AI. I'm curious.

Jordan Blackwell (14:14) No, it's a great question. And I know I was watching an interview where Jeff Bezos was talking about, know, kind of like the AI bubble that we might be in. And he was likening it to the internet bubble where I think he contrasted like industrial bubbles versus like financial bubbles where, know, 2007, 2008, that was a financial bubble. That's bad for everybody.

Whereas, you know, 99 2000, the dot com bubble, it was, it was a bubble, but it also laid out the foundation for the next generation of companies, right? You know, so all the fiber that was laid and, know, all this infrastructure that the internet needed to get to the next level. And yeah, I definitely see that where even if this is a bubble, it is not a fad in the same way that the internet

We did have the internet bubble, but the internet was not a fad. Obviously it's a core part of what we all do every day. And so I think moving forward, AI is just going to be a part of what we do every day, which means that there's, there's going to be a ton of opportunity. Like I just don't see a world where, you know, 10 years ago or 10 years from now, we look back and say, you're never that AI thing. Yeah. Whatever happened to that? Like, I just don't see that happening. Yeah. It's just, it's too powerful a tool. Obviously it's not, it's not a silver bullet for.

all the world's problems, but it's a powerful tool that, you know, a lot of people should be taking advantage of right now.

John Steele (15:47) Yes. I definitely think that there's a lot of room for us to differentiate ourselves, especially as we sort of build knowledge. I think that I'm seeing a lot of companies try and sort of lean into documentation, sort of find their secrets, put a stamp on it, and then sort of start.

you know, ravenously collecting data, writing things down, know, organizing, operationalizing things, because as AI grows, you know, as we both know, like it's only as good as the data that you can feed it and you to teach it and stuff like that. so a lot of companies are looking to do that. And that kind of leads into your Philly Tech Week presentation you've got coming up. So you got this event, ⁓ you know, it's going to be at Philly Tech Week, which is what date does it start again? I forget May 5th, I think.

Jordan Blackwell (16:41) May 4th? May 4th. So it's May 4th through the 8th is Philly Tech Week and then Ari Benzondeth.

John Steele (16:47) Great. Yeah. So, so first of all, I saw you were already at capacity. So congrats on that. But also, you know, ⁓ the event is called the role of AI and the great wealth transfer. And in the description, you talk about a number of things, but you position this really big topic here in our shared city of Philadelphia. And so I'm interested in, you know, why you chose this topic and why you chose to start in Philadelphia to talk about.

Jordan Blackwell (16:52) I appreciate it,

That's a great question. Well, like number one, like we're here, we're based in Philly. This is our city. I'm from here, Dee's from Portland, but like she's definitely Philadelphia now. And so last year was my first time at Philly Tech Week. And so, you know, was just getting Siah Labs off the ground and it was a fantastic entry point to.

just connections and understanding like what the Philly ecosystem looks like on the tech side. And it was just a huge resource. And I know that now and for a long time to come is going to be one of those staples in the Philly community. So props to what One Philadelphia, coded by and technically are doing with that because it's, it's fantastic. And so last year I was thinking, like, if I had the opportunity, I would definitely want to be a part of it.

you know, next year in a bigger capacity. And so when we saw that they were open to, you know, submissions for events and we were like, you know what, I think this makes sense. We would, you know, love to be a part of it. Now, in terms of why we chose the topic, I think it was just a confluence of like what we thought was prevalent right now. We had already had a few conversations regarding, like the silver tsunami and things like that.

And then obviously everybody's talking about AI. so we figured like we didn't want it to be an empty room. And so we figured, right, if we have AI and then you have wealth in the name, I'm sure that'll pique the interest of a couple of people. so, you know, putting that together and then, you know, from that time, just all the conversations that we've had, it's just made us realize that the conversation's even bigger than we had initially thought it would have been.

Right. And so we're excited. We almost think of it as just kind of a jumping off point where, you know, the conversation may start here, but this is something that's going to be a part of many conversations for, you know, the next decade or so as we are riding this, you know, great wave of the silver tsunami and things.

John Steele (19:20) You mentioned that Philadelphia had, you know, I think it was something akin to like there are a lot of family businesses here. There are a lot of like family led industries here. Is that something you learned in like researching the topic or researching? Clients for Siah Labs or?

Jordan Blackwell (19:33) Yeah, so we.

Right, right. Exactly. Exactly. So we were taking a look, I have to get the source, but they were saying anywhere between 10 and 12,000 in the greater Philadelphia area in terms of like small to big sized ⁓ businesses. And then on a national scale, they say that 92 % of small businesses end in closure. And so I'm sure that, you know, some, if not most of those are, they just didn't make it to profitability.

We know just from stories that we've heard and, just from the research that we have done, that there's a ton that have failed, you know, or I won't say failed that have closed and potentially closed prematurely. ⁓ potentially for some of the reasons like, ⁓ Deanna's grandparents, it's all right. It was just an owner dependent business. Everything had run through them. And so, you know, it, it, wins with them and, know, ⁓ without them, you

There's nowhere to go. so, yeah, that was where that stack kind of came from, the 10 to 12,000 small businesses. And then there are quite a few that are closing either because there's a lack of education regarding the different options outside of just closing its doors. Right. Or because they just didn't have the systems in place to take advantage of that.

John Steele (21:00) Yeah, definitely a conversation that a lot of startups, especially as the startup world has grown so exponentially in the last five to 10 years, we've seen a ton more conversation about how to prepare your company to get acquired, how to do succession planning, like you say, so that we're not seeing these businesses close, but sort of have a happier ending. And I'm sure some of these things wind down on their own volition. Some of them are not profitable, as you mentioned, but I think, yeah, I think there are a number of companies that, you

Jordan Blackwell (21:21) Right.

John Steele (21:30) don't know how to do that and don't really know where to go and may end up, you know, not getting great advice about stuff like that. And, and, and so, yeah, you in your, in the, in the literature around this talk, you guys mentioned the, the importance of capturing institutional knowledge before the silver tsunami that you mentioned earlier, sees older owners retiring. So can you talk about that risk in, in a little bit more detail? Just talk about why AI especially is uniquely suited to protect against.

the risk of not capturing that institutional knowledge for yourself. And how does that bleed into founders building companies that maybe can get acquired or maybe can be passed on to the next generation? you know, what is that all, how does that all tie together?

Jordan Blackwell (22:14) Yeah, no, it was a great question. And I think it goes back to what you were saying where it was, you know, AI is only as good as the data and the context that you give it. But if you do give it the correct context, it is an extremely powerful tool, which enables things that before just were not possible. You know, where before, if you have it, you know, connected to your various data sources and also kind of the reasoning behind, you know, certain decisions that you make.

Now it does allow, you know, someone who's not the owner to ask a question or to understand, we're, giving this discount because three months ago you had this conversation, you know, that said, you know, X, Y, and Z. Yeah. Where before it was either impossible to do, or it would take a lot of manual effort to do so. And so the, with that kind of how I like to think about AI is where, you know, the internet

If you think of it like us getting a book of the world's knowledge and then AI is kind of like us getting a tutor that has read that book. Right. And so that, that it just lowers the barrier to entry to learning to, you know, you know, learning new things, understanding what's going on. They're, just extremely powerful tools. And the, on the business side, it's kind of the same thing. You need to build that book that AI as the tutor can learn. Right. Yeah.

Because, know, AI just learning about, you know, the capital of France, isn't going to help when you're trying to understand why, you know, this month's revenue is lower than the previous quarter. And so that's kind of our job. Our job is to build that book and then create that tutor, you know, based off of that, that book of knowledge. And you have the actual kind of data sources, the orders, the sales, the deals, the, you know, the appointments, all that good stuff.

But then you have the other half of that, which is almost the invisible knowledge, which is the tacit knowledge, the knowledge that is in the owner or the operator's head, where it's like, Jeffrey knows everything about that. So if, if he leaves or if he's sick, like we're in big trouble, like we want to, you know, kind of get rid of those types of situations where, you know, it's the platform understands what's going on, which frees up, you know, not only

Jeffrey, but everybody else in the company to focus on what they do best. Yeah. So that's kind of how I think about, you know, AI and this, this knowledge capture.

John Steele (24:46) Yeah, what and what do you think, you like you mentioned risk, for example, what do you think the big risk is to people not capturing this institutional knowledge? mean, so is it is it just to feed AI or is there is there a greater risk outside of that, you know, for folks that are trying to, like I say, like, you know, make themselves a little short up their operations a little bit more. You know, I was curious about the term risk and just how

How does it impair all companies if they are not able to sort of do this work?

Jordan Blackwell (25:18) Yeah,

you're making decisions on incomplete knowledge, right? You have, it's almost, you know, two sides of the same coin where you do want to see, you know, what your orders look like this quarter. But you also want to know that, Hey, like every second quarter, you know, every Q2, we have a small dip. And, you know, the owner knows that because for 20 years, they've been doing the same thing. They understand why. And it's because,

The vendor here likes to take their time in Q2, but they always say that they'll make it up in Q3. And so now without that tacit knowledge captured, now you're just saying Q2 is an issue. Now we need to make some drastic decision when in actuality, now this is the natural course of how this particular business runs. And so I just think it's when you're making decisions off of incomplete knowledge, that is, that's the big risk. That's the big risk.

for you as the, if you're an incoming business owner or if you're the existing owner and you are looking to, you know, pass that off, they say that you could have as much as double your exit multiple when you're selling a company, if it's systems led versus an owner dependent, you know, company. so that right there is a very tangible risk of you're not getting the valuation that you actually deserve because you don't have the systems in place, you know, to, to warrant that.

John Steele (26:48) Yeah. So we started to talk a little bit about AI and I know it's very important to what Siah Labs does, but you know, many, many people out there have a deep distrust of AI. So a recent survey I saw said only 5 % of Americans say they trust AI a lot. you know, there was another study that was released that said that Google AI overviews gave incorrect answers up to 15 % of the time.

And that may seem relatively low for new-ish technology, but when you consider the insane amount of searches that Google gets, you've seen companies and people defamed through these AI overviews. So there are certainly some issues that are impacting trust with AI. And I'm curious how that, you know, how you guys talk to business owners, especially older business owners or folks that have been in the game a long time.

and are looking to shore up, looking to do like succession planning, looking to, you know, start building legacy sort of. mean, that's a lot for them to trust in. And I'm sure when you mention AI, at least some portion of them have to be like, I'm not going anywhere near that until they figure some stuff out. So like, how do you sort of talk about it and sort of try and, I don't know, rebuild some of that trust with the folks that you're speaking with?

Jordan Blackwell (28:08) Yeah. I mean, that is an excellent question. I think one of the big things is what we were talking about before where AI is not like the tool to end all tools. It's not the end all be all. ⁓ and I think that definitely needs to be instilled from the very beginning, like from the first conversation.

John Steele (28:28) Plus

you ask the AI, the owners of the AI companies, then it's...

Jordan Blackwell (28:31) Yeah,

no, no, right, right. Then like none of us are having jobs.

John Steele (28:34) really

is, it's the god of gods or whatever they've been saying about it.

Jordan Blackwell (28:38) Exactly. No. ⁓ But I know that I wrote a post on this talking about more tactically like deterministic workflows because AI is what they call like probabilistic or stochastic where if for every input you are not guaranteed to get the same output. And especially for business, that's something that like it doesn't really fly. Like you want to be able to say, Hey,

If I'm putting this in, I'm getting that out. ⁓ and that's where like just old school code comes in where there are certain things and you have to be able to delineate what pieces are okay to hand off to AI 100%. What pieces are okay to allow the AI to do the bulk and majority of versus what are the things where you need to have that human in the loop? ⁓ and maybe have AI as a, like an assistant to help you.

along that path, but there's definitely a spectrum and it depends on kind of what the workflow is and kind of what it's touching where it depends on that. But that's all part of, I know we've been talking a lot about the product side of it, but there's a big consultative part of the business where we're in partnership with these businesses and we have to, you know, talk about what it is and what it is not, because there are some people that are on both extremes where

I will not touch this at all, or I will trust this with every single password and account that I have. And we need to bring both of those back to homeostasis where it's like, no, it's like, it's, it's very powerful, but it is not everything and it cannot be trusted blindly. ⁓ and yeah, I think that's a big part of the service component of Siah Labs to go along with the product component.

John Steele (30:30) Yeah, that, that makes sense. and you know, I wanted to ask you about this too. You know, there have been a number of AI tools that have either shuttered or come under fire for different things. There have been some tools that, you know, started out and then sort of, you know, not obsolete, but sort of just like secondary and haven't really caught on and things like that. How should owners and founders, you know, prepare?

for the volatility of AI. So I think a lot of times you see founders trying to incorporate AI into their businesses. But I mean, if you make AI central to like big workflows in your company and then that AI company like goes belly up or, you know, gets acquired and they change the way they do things, I mean, it could really throw a wrench into things that you're doing. So I'm curious how you're counseling your clients ⁓ to sort of protect themselves again.

Jordan Blackwell (31:26) Yeah, no, that's a great question. So there are a couple of things. So one is depending on where they are in terms of wanting to learn the fundamentals of it, because it's one thing to go with a company where essentially like for Siah Labs, every one of our customers has their own unique instance. And so that means that it's not the Siah Labs platform and then your row number 34, you know, of customers in this platform.

Everybody has their own unique instance, its own app with its own database. there's no risk of like cross-contamination of data or anything like that. What that also allows for is like kind of capping the downside. Cause say something did happen to Siah Labs and we're like, Hey, we actually, you know, can't go on anymore. We're at a point now where each one of our customers could take the app, you know, along their data, along with the code.

and run it, you know, themselves, right? And it's all on, you know, a popular framework that, you know, many people are familiar with. And so from that end, in terms of capping the downside, that's where, you know, Siah Labs is uniquely positioned where they won't really be caught, you know, without a solution.

John Steele (32:40) Mm-hmm.

Jordan Blackwell (32:42) I will say, I guess the core of what I was trying to get at is if you are going with a custom solution or solution that you're building yourself from the ground up, then so long as Google doesn't go out of business or you know, the large providers, ⁓ then you're in pretty good shape, which I don't think, you know, in the next couple of years, that's going to happen. Right. But if it's something where you're abstracting away everything to where you are with

You know, you're, calling an API of some third party company that, you know, just came about a couple months ago. That's when that risk really does happen and show its face. And so I think for capping that downside, it's being as close to the source as possible. Ideally it's, you know, building those custom workflows where if they decide, you know, we're no longer wanting to go with open AI models, we want anthropic or we want open weights models, then it's.

kind of a smooth change out as opposed to now I need to find a totally new company, you know, that deals with these other providers. And so yeah, getting as close to the source as possible allows that flexibility.

John Steele (33:52) Definitely, definitely. You and I both like to share our experiences on LinkedIn, talk about AI experiments that we've worked on, things like that. What has been your most successful AI experiment to date, whether it be professional or personal? I know we both like playing around with these tools and sort of talking about, you know, what they can do, but what's something that you've built using AI or done with AI that you felt like was a home run? You just were like really psyched about it.

Jordan Blackwell (34:20) All right. So there, there are a couple of things. So I'll say one has to do with my four year old son Josiah. And so, you know, Josiah obviously loves spending time with him, love, you know, doing different activities and kind of trying to introduce him to, you know, software technology, AI, not anything too in-depth. But one of the things that he really gets into is, ⁓ like different videos. And so.

What we've done is we've been able to draw out different pictures, you know, by hand, and then we turn that into actual like generated videos or, you know, we have like different adventures, you know, kind of creating like a book per se. And just seeing like him be able to take something, you know, that he created and then turn that into like a full quote unquote production, ⁓ has been, has been a ton of fun.

John Steele (35:14) Yeah, they've been doing some of that at my daughter's school too. yeah. And talk and build things with AI and stuff. It's pretty cool to hear her talk.

Jordan Blackwell (35:19) It's fantastic.

Yeah.

I just, I can't imagine what their generation is going to be like, because for us, this is all new. This is like, oh my goodness. For them, it's going to be like, oh, it's AI. Like I was born with that, you know?

John Steele (35:39) Exactly, the digital natives except the AI versions.

Jordan Blackwell (35:42) Yeah,

exactly. But the other one was we started building like a Siah Labs for Siah Labs and started to dog food our own products. Yeah. And it was, it's funny because it started like with no AI, but I just created essentially just like a little note taker where whenever I had a thought, whenever I had, know, an idea, an article that I liked, I just kind of put it in and it would just accumulate.

John Steele (35:52) Yeah, yeah, yeah.

Jordan Blackwell (36:10) And I did that for months because I knew, like I knew that the whole context is key piece of it. So even though there were no AI features or anything like that, I knew that if I was, you know, kind of feeding it the correct context over the course of, know, a year or so, it would eventually come in handy.

John Steele (36:29) Tell

the AI to reference that. just from a setting that up perspective. Like if somebody watching this is not familiar with how to do this, like how would you put material that sort of shows your way of looking at the world? How do you sort of like tell AI to like reference that as a sort of point of like writing about you or something, writing as you or something like that?

Jordan Blackwell (36:53) For

sure, for sure. ⁓ So the, the core of it, if you're thinking about like an AI agent, like what an AI agent really is, is a large language model. It is a set of tools and then it's a loop that it controls. Right. And so as opposed to a workflow where you're controlling like when it starts and when it stops, the actual LLM is deciding, okay, either I need more information or I'm done and I give the, the result to the user.

And so in that case, what you would do is you would have an LLM. So you'd have like a open AI, Gemini, Claude, whatever it may be. You have that. And then the tool piece is where you get to have a lot of fun, where you can have a search notes tool where it's deterministic code. it's, Hey, you, put in, you know, previous seven days and it returns all the notes from the previous seven days. It inserts that into the context window. the memory.

of the, ⁓ the LLM. And then once the, the model decides that it has enough information to answer the question or, know, whatever the case may be, then it takes all that information. does its thing. And then it spouts out the answer. ⁓

John Steele (38:08) That's

cool. That almost makes your thoughts and ideas and things that you find online, like a digital asset management system almost, where it's just referencing that like an automation used to access a dam, but now you're doing it with AI. So it's influencing its thought process as well, not just using the pieces.

Jordan Blackwell (38:28) Exactly. Exactly. You can go so deep with it where, you know, obviously there are things called agent skills, where they're marked down files where essentially you're saying, okay, if you ever come across, you know, if Jordan ever asks about like the house here, here's a markdown file, a document that has all the thoughts, all the key figures and whatnot about the house and use that as context. And so if I'm talking about, you know, playing the piano or how to do this.

And then later on the conversation, I'm like, wait a second, like, when did we replace that roof? Like then it can go to, you know, that, that skill file or that markdown file, it get that information. And so, yeah, it really is. The, those tools and then what you put in as in that tool set, it really has everything to do with the deterministic workflows to ensure that you're getting the same result every time. And then two, it's about the context.

And so that's, you know, kind of coming full circle, you know, whether it be personal wise, you know, or business, you need to ensure that you have that book of knowledge from which the agent, the workflow can draw the necessary information. Yeah. And so, yeah, when you get it right, it is, it's a fun time.

John Steele (39:45) Yeah, I've been really interested in the memory piece of, Claude and I've been really interested. I use a lot of them, but I, but I use Claude primarily and I have been really interested in the memory piece. It started to remember things about me from, from months ago and it'll add that as context into, you know, responses that it gives me. And especially in the chat bot, but even in like coworking and code sometimes it'll be like,

Did you want me to do this? Because I remember you said this other thing and it's like, make connections is insane. So the more you can control that, think, think a lot of times people's fear about AI is that kind of stuff, but the more context you give it, the more control you have over what it is. You can be like, do this, don't do this. There was somebody recently about the planning feature in, in cloud and see how it's planning to.

Jordan Blackwell (40:30) Right.

John Steele (40:37) attack a project and you can say do that part don't do this other part I don't want you to And that's been really powerful for me too.

Jordan Blackwell (40:41) Right, right.

Cause also with the tools, thing is like sometimes there are certain things where you don't want to have to rely on just telling the model not to do it. Cause I use this analogy where it's like, I love my son, you know, I know he's smart and he wants to keep himself safe. But like me telling him not to open the car door is not what makes me safe or what makes me comfortable knowing that the car door is not going to open. It's because I have a child lock on the car door.

So that no matter what happens, if he goes rogue and he's like, no, I'm trying to get out. know that, you know, car door is not opening. And that's the difference between, you know, probabilistic of me just saying, Hey, please do not open this door versus deterministic. Like no matter what happens, that door is not opening. so, yeah, that's all part of, you know, trying to, you know, just keep that balance, you know, with AI. Yeah. It's a fun time to build, fun time to learn and figure all that stuff out.

John Steele (41:41) That's a great analogy. So I have one final question. ask this to all our guests. What is a tool integration workflow or automation that you can't live without? or personal? Doesn't matter.

Jordan Blackwell (41:54) Yeah.

Yeah. You know, here's actually, I don't know if it's, can't live without, but it has been huge for Deanna and I, ⁓ the last, you know, couple months, which is we have lots of meetings when, know, we're like brainstorming and stuff like that. And before, like I'd be in front of my, you laptop or she'd be in front of her laptop and we're trying to get all the good tidbits out of it and whatnot.

What we've decided or what we try to do is to just record ourselves and then, you know, pass that through AI to give the summary, to extract all the important information. And it just allows us to focus solely on the content of the conversation as opposed to, ⁓ you know, ⁓ devoting too much energy towards the, wait, wait, wait, wait, say that one more time. I don't think I got that the first time. Like that was really good. no. ⁓ and so that's been huge just in terms of.

John Steele (42:45) Yeah, yeah,

Jordan Blackwell (42:50) you know, the throughput and the, ⁓ like been able to, you know, sketch out blog posts from that in much ⁓ less time and, much easier because now we're just talking about what we want to say as opposed to, you know, staring in front of a blank document. So yeah, I'd say using AI to, to summarize those, ⁓ audio, files. That's a big one as of late.

John Steele (43:17) Nice. All right. Well, hey, thanks a lot for joining today. I really appreciate it. And I'm really excited about your Philly Tech Week presentations. Fascinating topics. Really interesting. And I've thought about AI's capabilities in sort of backing up an institutional knowledge, but I've definitely talked about in our newsletter, Funnel Vision newsletter about the importance of doing that for the very reasons you said, that you have to feed AI this information and using it that way.

Jordan Blackwell (43:21) Thanks for having me. This is fun.

That's a movie.

John Steele (43:46) But the interest in sort of companies trying to pass on to the next generation is super fascinating topic.

Jordan Blackwell (43:53) Yeah, we're excited. We're excited. Happy to be there. No, for sure, for sure. And if you're around, definitely come through. I'd love to have you. Perfect. All right.

John Steele (43:56) that you got in.

Wait, I will try and do that.

Jordan Blackwell, thank you very much for joining us today. I appreciate it. All right, bye bye.

Jordan Blackwell (44:07) Thank you John, talk to you.