Sean Lorenz
===

Jeff: [00:00:00] Hey, Sean, great to have you on man. Stoked to talk with you today.

Sean: Good to talk with you. How are you doing?

Jeff: I'm doing well. I'm doing well. Holidays ended and I think we're in a good spot now. 

How would your holidays go?

Sean: It was great. Other than having to drive with. Two kids, two animals to Ohio. It was pretty good.

Jeff: I've done the Ohio drive. It's a long drive and I did not do a two kids and two animals. So good on you, but cool. But so let's jump right in. Now you've been doing AI since before AI was cool, right? Like you, you are

Sean: I guess that's true.

Jeff: you're old school AI. You did it, you're the hipster of AI.

You got your PhD back in 2012 in AI and neural networks and then wasted no time founded Neurala into tech stars. Like you want to tell us a little bit about that? How'd that kind of go? How'd that happen?

Sean: Yeah, sure. Few professors and colleagues at Boston university, they had already had this idea to take at the time there was something called neuromorphic computing, which still is in its [00:01:00] infancy. And there's a lot of interesting work being happened there where a different kind of compute chip other than a CPU and A GPU that uses how the brain computes with spikes to create chips on silicon.

It's really fascinating stuff. And they wanted to create this company to basically be more biologically inspired neural networks as opposed to the more traditional artificial neural networks that are. Based on back propagation and honestly, the backbone of what most of AI is built on today, right?

And it was a very different approach and after I finished my PhD, we applied to Techstars out of It was a NASA STTR grant that we had gotten. I literally copied and pasted. Don't tell Techstars that I'm here. But uh, I did it.

Jeff: we're outside the statute of limitations now.

Sean: That's right. I copied and pasted parts of the STTR we wrote and put it into a Techstars application and they loved it.

We were really [00:02:00] doing something very different building a navigation system for spatial navigation. Mars Rover was the intent, to essentially use a rat's hippocampus as the model to put it on a Roomba. And that's literally what we did Roomba without the vacuum cleaner part.

And that was the basis of our entire technological advantage

Jeff: Nice. So yeah, you guys didn't have big designs at all. It's, some people want to build a new way to pay on the internet. And your company was focused on let's navigate Mars like a rat would.

Sean: a rat. Yeah.

Jeff: yeah let's put a Roomba and a rat on Mars. I would say that those are big aspirations but no, in reality, that's pretty awesome.

That's, that is the world changing stuff that, we all dreamed of being a part of when we were all joined startups. So what happened in the end? Like where'd we end up here?

Sean: Yeah. So with Norala, you know, Techstars Really, I think helped catapult the company. And , during it, we were a technology in search of a problem. Really, like you were saying, we had [00:03:00] this great idea, very big idea. And, we were looking into, I don't know if you remember from a brief moment, there were those little sticks with an iPad on top doing telepresence robotics

Jeff: Yeah, we had one at a company I was at.

Sean: So we were trying to make that smarter. But . The market was really small and it was already dying and honestly it just was too little too late. And so we ended up the getting a phase two for that NASA contract that was more focused on actually pivoted a little bit more into spatial navigation with drones.

So automating drone collision avoidance, making sure that drones don't hit each other. So in that case, we used another. mammalian brain model actually of how birds do collision avoidance using optic flow really fascinating stuff. And that has been the backbone of Neurala. It takes a while to get funding for something so out there and next [00:04:00] level generation kind of tech that, At the same time, I was really getting interested in the internet of things.

And I got lured into that world at log me. And they had just acquired a patch bay out of London and they turned it into Xively. Building a really amazing internet of things platform. So I pivoted out of the early stage startup into a large enterprise, SaaS kind of environment down in Boston.

Jeff: That's what I'm going to say is it's funny because, despite kind of a lack of there wasn't a successful outcome with Neurala, but it really launched you from, a PhD kind of academia, which can be a little pigeonholing in tech sometimes into a very, accomplished career and product where you've really done a lot of cool things, but you went over to to log me in.

And I remember Xively because I had a good friend at the company at the time and the CEO was talking big about it and it was going to be, the new future. Of, not just logged in, but the future of how things in your house operated. Um, you went [00:05:00] there and so , what drew you into IOT like that seems a little bit far from let's put a Roomba with a rat's brain on Mars.

Sean: Actually, it's funny cause Neurala is still doing great there. And they pivoted into more a visual inspection which is all, basically how do you have smarter inspection of, you know, manufacturing floor? And honestly, I feel like that was all robotics. I just see it as sensors.

It's all sensors. And so the internet of things in my mind is just an extension of what we're learning in neural networks at our program at Boston University is, how do you take a lot of sensor inputs and process it cognitively to create an output? Right? And so at LogMeIn, I saw the future Of saying, look, all these different smart devices are just collecting sensor data, different modalities.

How do you stitch all that together? At Neural, it was robots, but at LogMeIn, it was anything. Temperature, humidity we were looking at industries across [00:06:00] all kinds of things, building smarter diapers. A sensor in diapers, that was an actual pitch. I saw the value of moving from not just one kind of sensor, but how do you combine all these different sensors together and not just connect it, but how do you actually analyze it and then turn that into actionable intelligence and build a whole platform?

Jeff: Let's back up a little bit. Because it, I find your kind of general path really intriguing because you, as much as you, you got the PhD in AI, you did not start in the hard sciences. I think you originally got your undergrad in philosophy.

Is that right with? Yeah, and I think we, so we talked earlier and this one kind of killed me because, a little bit, I take a little offense here, but not much, basically, I think your take on that was it set you up to either deliver pizzas or work in marketing. So I'm not gonna, I'm going to try not to take that one too personal,

Sean: Don't take offense. Hey, I was in marketing for a while and I

Jeff: Yeah. Now that we all try, one day I hope to get out. But it [00:07:00] seems like you made it through and you proceeded and you then went from marketing to you proceed to help found the independent film festival of Boston Like what is your life here?

I think this is its 22nd year going into now. So that's amazing. 

I mean, So you get through your art phase You go and get your phd in ai you help found nirala and then you end up at xively now If I remember correctly, the team over at logmean talked, you know was really excited about xively It was going to be the future.

Don't think it worked out that way. But maybe walk us through, , what went on over there and how it came up and ultimately where it went.

Sean: So Xivli was a really fascinating case study in finding out how to build a platform and knowing how much marketing at the time I was on more of the product marketing side is a little bit of strategy, talking back and forth between product management and product marketing. At the, at that point, everything was just straight connection.

How can you [00:08:00] use MQTT and these different kinds of protocols to just connect something to a little Raspberry Pi or some sort of device and productize it. Um, and, Xively saw, we were really trying to push the whole thing of, we said it was like connect, manage, engage, I think was our, what we came up with.

And We wanted to get ahead of any competitors by saying, look, yeah, we can do the whole thing, everything from not just connecting to the device and partnering up with TI and these different boards, but how can we take that further and say how do you productize this, what's the value of your product?

So honestly, we were helping them build business plans for their connected device to try to get them to buy ours. But then we started really talking up well, yeah, we can do all the analysis and we're going to have this big, full holistic platform. And yes, I think we were going to get there, but really we needed to get the connection part, locked in, but, just saying, yeah, we have MQTT to connect wasn't as sexy as [00:09:00] saying, yeah, we're going to do AI on your connected windmill, right?

I think that's where there was a little bit of a disconnect as we got out, the marketing message was a couple of years ahead of where, the product actually was at the time. So I think that was the part of why we didn't get as large at LogMeIn with as Ivy as we could have.

Jeff: Yeah. It's always that needle to thread, right? You stay too far behind, you sell behind your tech and someone's going to out innovate you, or you're always going to have the competitor. Who's a little bit more pushing boundaries and everyone kind of looks at them as the market leader in tech, you get too far over your tips.

No one wants the thing you have, they want what you're talking about, which is, Five years out if you're doing it too far and that's the danger. 

Sean: And as a product lead, I think it's really important to really work closely with your marketing sales team to say, look, this is where we're at, but helping them see the near term roadmap and say, I think it's okay to sell a little bit of vision and say, look okay, this is great.

We have this now, but [00:10:00] let's get a contractor, some NDA or something in place where we will help you see. That this is the next year coming. So that keeps them almost hooked, but at that point you're saying this is the reality. We. Really hope to get to this within the, the timeframe of your contract.

Jeff: It's almost two separate sales processes. You have to sell them on we can solve this need of yours right now And here's where how we are going to understand and solve your future problems as well But there are two separate things if you just do the first the second one good luck you'll always be having a customer who's ready to sign in two years 

Sean: Yeah, exactly. That's

Jeff: and then So Xly, you kinda went through a great case study, right?

And you need to match up timing with how you sell and how you go. But then , you continued on your iot play to a company called Center. Is it?

Sean: Yeah, that's right.

Jeff: And, 

explain what Center does please. 'cause I think it's a little less I didn't know about this one beforehand. 

Sean: So center was out of log me and I really wanted to get back to [00:11:00] my. Mix of neuroscience and AI roots and wanted to leverage smart home devices, off the shelf devices to put that together with that promise of AI that I was hoping we would eventually get to with Sively.

Or a specific use case, right? I wanted to solve a specific problem, not be a horizontal platform. And so with center, what we're trying to do and actually what we started to create was the ability to take remote passive sensing of location monitoring sensors, as well as bed mat, eight, all these new kind of sensor devices as smart home devices were just coming on the market.

It had early. Beta prototypes. And so we were working with them to say, Hey, give us your early access to your API APIs. And hopefully if we can sell this, you're going to sell this as well, right? If people buy ours, they're going to buy your bed connected, mad or your smart things, emotion sensors, right?

We packaged it up and we were trying to find [00:12:00] solutions for remote patient monitoring for seniors in particular. We could accurately tell urinary tract infections just by putting a sensor on a in your bedroom and connected to your bathroom to say, okay, at what time did you get up at night?

What. What were your patterns like? What is your sleep like? There's all these kinds of triggers that could also be used for delirium and signs of dementia. So that was really the thinking. And I was speaking at dozens of conferences at the time, digital health startup to watch.

But when we were going up for fundraising, no one wanted to be a lead because of product market fit. It's okay, technology in search of a problem again. And they said, look, this is a great idea, but who's going to pay, right? You're not going to, there's no CPT codes at the time in 2014, 15 for a remote patient monitoring.

Obviously with COVID, a lot of that changed. It's a different story, we are too ahead of the curve and there was no payment model. Um,

Jeff: You just explain real [00:13:00] quick for people maybe what CPT codes

are as well, please? 

Sean: Yeah. So in the healthcare system, you have a fee for service code. So every time you go to a doctor or any time you interact with your healthcare system, it's a code.

It's a reimbursement code that says, hey, you get 200 for every annual wellness visit or every visit for this kind of chronic care. condition, you get 100 bucks, right? And so a lot of the software as a medical device world is built around trying to find those codes that you can attach to get a piece of that when you sell to a certain part of the health care system.

Now, obviously, the future is more value based care, which is not using those CPT codes. Everything there is about preventative care, but the U. S. healthcare system has been really slow to adopt that in full force. So trying to build a digital health startup around a value based care model, it's tough. You have to have enough money in the bank and a value proposition that [00:14:00] is, airtight to really handle that and leverage and manage that storm of, if you have to go for FDA.

Clearance which could take years to prove that out

Jeff: I feel like there are so many lessons to be taken from this, few year period of your experience here where, at center you talked about you were out and talking at all the conferences and getting a lot of attention and traction. And I think no matter what industry you've been in, you've, we've all seen that competitor or maybe that vendor that we know, and you're just looking at them going They're killing it.

Why can't I do that? And we all get maybe a little jealous of it. And then what you don't know behind the scenes is there's so much more to having a successful business than just product market fit, right? You need channel market fit. The market needs to be there. You need to have the need lined up with not just the need and the product, but with also how does someone buy this thing?

And the world is, sloppy with [00:15:00] the people and the companies that have come in and made a lot of noise and had great, PR or, had a, in your guy's case, a great product that really did something valuable, but just not the means to monetize it because the channel to go to market just hasn't caught up.

And then, similarly, if you look back to Xively it's a great product and a great need that was going to do stuff. But if you don't, how you go to market. Matters as much as the product, right? You need to 

message correctly. And that's where you need to have the discipline around messaging and the discipline around not just either right product, but I have the right channel to go to market.

And these are all the components that build a, solid sustaining business But it's so easy to look at some of these companies and go I wish I had that thing. And you never know what's going on behind the scenes.

Sean: exactly. You don't. And so it is actually something I've definitely learned out of center is how much of this. You don't know what's under the books. You don't know how much is just marketing versus actual revenue generation. And that's [00:16:00] really what matters when you're running a business. And I think that was the biggest thing I learned out of both, Xively and a bunch of my early startups that I started in my career was, Trying to do more homework up front in terms of, okay, yes, there might be a really cool technology we can apply. There might be product market fit, but you have to have that match to your point, with all the other components and does the business, is there a business model?

And it's not just going to a customer and say, Hey, do you like this? They're always going to tell you yes to your face. But when you say, all right, this is the cost. Will you be a pilot? And they say, Ooh, I don't know. That should be telling, as a product manager, if you hear that should be an alarm bell.

And then you ask them, okay what gets you over that hurdle? And if so, that's great. If not, why? And then that should help you inform that before you start building anything. And that's where I think we went wrong with center is we built out this thing [00:17:00] first. And then I said, Hey, you want it, instead of saying, doing the opposite.

Jeff: Yeah, exactly. No, that's so key is a lot of people are going to say, yes, people are nice, but also it does sound like a good idea no matter what. But once you get into the messiness of, okay, I do want, but how am I going to bill it? How am I going to pay for my subscription of your thing that you've built?

Then once you do that, how do you get, 10 more, 20 more, 50 more people to do that? It's, you need all that compelling reason that it's all going to work. 

Sean: In healthcare, especially with the healthcare industry,

Jeff: it takes so long.

Sean: It takes so long and they don't want some, your fancy app or your fancy portal, right? What they want is a, is just one field that gets updated in their EHR system, right? Is if you can show that you can seamlessly integrate with Epic or Cerner or one of these and say, Hey, Yeah.

We see your diabetes management platform, it's a little device, this works really well. And Oh, by the way, it automatically hooks right into your [00:18:00] Epic system and you don't have to worry about it. And we automate the billing process. Boom. Then you start to have a conversation.

Jeff: Exactly. And then But, there is a light at the end of this tunnel. Moved on from Center and ended up at Formlabs, another, great Boston company really pushing the bounds of 3D printing. And this, I feel like, is just that great case study of, product meets need, meets the right go to market channel.

And things went really well there. To the point where I think you, you told me at one point, and you're gonna have to tell the story better, where you guys were at a conference and literally had to go get like one of those square devices to plug into your phone so you could sell stuff there because there was so much demand.

Sean: Yeah, our events team came back from I think it was like a dent. They were the man, they did so many conferences. But they came back from a dental conference, this was, the early days of their dental market , they had a couple of resins that were fit for dentists and, they went to this one and They had dentists just coming up to them saying, [00:19:00] I love it.

Can I buy it? And they're like, wait, what? They weren't prepared. And so the ad, they had to go out, send somebody out to get striper, one of those. And yeah, square, and so they could swipe credit cards and they were selling them, right there on the floor at a conference. I've never seen anything like that before.

And that to me shows you they're onto something. And so really. That following year, we ramped up our dental resin market just by leaps and bounds. It was just a great option for product market fit. It just kept growing and growing, you know, into dentures and other types of items where, dentistry is a perfect example of where if you can do it quickly and cheaply and safely to, you know, customize your teeth.

Everybody's mouth is different, earbuds, teeth, sunglasses. They were perfect use cases and people wanted them especially in dentistry. So I think, 3d printing has revolutionized. The dentistry industry.

Jeff: [00:20:00] It seems like that is one where it was just an overnight success, but I'm sure it was not that successful. Easy. What was the work that went into kind of getting you there, right? What was the, 10 years behind becoming an overnight success kind of thing? How did you guys identify that market and get in there and figure out what it needed and make sure you had the right thing at the right time with the right channel.

Sean: With Formlabs, they had already been just absolutely, crushing it in more of the prosumer market and, they had started already, when I got there a little bit, the dental, but really it was just listening to the dentists in that market, as well as in, the healthcare market in general, that was another new field.

And, we were really trying to go from a. prosumer kind of market into more straight B2B applications. We are looking at, partnering with SAP for manufacturing. And that ended up being a new industry where we were looking at jigs and fixtures and trying to figure out. Where in healthcare, [00:21:00] in hospitals at Northwell in Long Island, I went out, they created this audit, we created this automated system where you could have eight, form lab three printers printing parts right there and give them up to a surgeon right before the surgery or after the surgery.

Stuff like that just was impossible before. So we just literally just listened to our users in these more business use cases and said, What do you need? And we just watched them, that I think a good product person goes out into the field. And, I went to factories, I went to hospitals, I went to Dennis office and said, and we just figured it out and said, okay, let's create more residents that, have different applications with different properties they needed.

So it actually was overnight. It was really amazing.

Jeff: But that's the right way where, you did the work of building the machines, having the use case, you understood your users. And as you rolled out resins and all that kind of things, you knew that Hey, you knew the demand was there. Cause people were trying to buy it from you when you didn't even have a [00:22:00] means to take payment.

You. Knew that you had the channel to go. You weren't, getting in the way of some weird billing procedures or anything like that. And that's where everything the skids are greased for lack of a better term in

Sean: Yeah, exactly. Yeah. That was, that's a perfect analogy.

Jeff: so cool. And now you're at Altoida which I do love your just overall thesis.

It seems like your career of. Let's take this thing, we all know is AI and let's continue to try to improve, healthcare outcomes with it because it's such a meaningful place where we can make a lot of improvements. , and, fundamentally, all this stuff hasn't changed that much in a long time.

So it's good to see kind of people focusing on how do we move this world forward in a way that is, is probably a little bit slower. So good to have dedicated

Sean: A lot slower.

Jeff: Yeah. Yeah. Yeah. And so at Altoida, I mean, a few of the things you mentioned to me just before we jumped on this were mind boggling and how you were doing it.

Like 10 minute diagnostics to, to [00:23:00] predict dementia stuff you're doing that people can do at home that used to have to go do, I think like hours long tests to get the same results.

Sean: Yeah. Yeah. That's the problem with the neurology world is, neuropsychological exams. It could take especially for a neurology visit, nine months to a year is oftentimes the case to get a appointment for a dementia screening. And those are typically three to four hour neuropsychological batteries and exams.

And, if you're a 70 year old. Person who might already be showing signs of dementia waiting a year is just going to be too late Also, you're not getting their baseline When you have to have them come out and do three to four hours of exhaustive examination It's just not the way to do things.

Altoida really is a game changer and saying okay how can we put this just on this, and we were able to do it at home as well as, in an office where it's [00:24:00] just a bunch of tasks where we have you tap things, we have you draw circles, and then we use the augmented reality toolkit on the Apple iPad to have them place virtual objects in a room.

And then we ask them where those objects were so we can look at all the accelerometer data, we can look at all these different types of data native on the device, and we just throw it into a machine learning algorithm. And then we get ground truth, because that's, The key is the amount of data you can amass and compare it to ground truth.

So a lot of money goes into these kinds of healthcare studies where you have to validate, okay, they take a 10 minute assessment, which that's the easy part. The hard part is honestly getting all the ground truth to say, okay, do they have mild cognitive impairment? We have to look at getting an MRI scan, get their blood drawn do the Tests that exist and then say, okay, how confident are we that we predicted [00:25:00] this correctly?

And that's why a lot of, healthcare can't Break things fast and break it and move fast kind of mentality is because these are people's lives. And you have to make sure you're extremely careful when you're getting these sensitivity curves and scores to predict whether someone's life might be changed by , a one that says you have, A high likelihood of dementia

Jeff: It's one thing if you move fast and break someone's ability to post on social media, it's 

Sean: or their Netflix recommendation, right?

Jeff: and break someone's sense of am I going to, have cognitive impairment for the rest of my life? And do I need to prepare for this life changing thing?

Little bit different weight.

Sean: yeah, a little different. And the nice thing is once you have this very personalized set of scores and cognitive subdomain scores, you can build treatment plans that are tailored to that, and that's where I see the future of AI and medicine going It's early intervention through diagnosis into treatment. And that's the part that I think is still early, but [00:26:00] personalized treatment across cardiology mammography, we're seeing it happen neurology, all the different, Parts within, the healthcare system, when we can start tailoring custom treatments, instead of getting a 10 pages of, okay, go home with this generic set of instructions that nobody reads, no, we should be able to go home with a personalized treatment plan based on exactly your kind of thumbprint.

So that's where I see things going. 

Jeff: To the point you made though it's not just the tailoring, it's getting it way ahead of when we can do it now and historically it's, we know with dementia, early treatment and early kind of intervention is a huge lever to helping this person for the rest of their life.

But AI, it seems like has the capability to open our ability to do that in many cases. There's so much data in the world that is, obscured by just the fact that it's life and there's just so much going [00:27:00] on. And yes, if you had a sensor on you and all these things all the time and enough computational power to parse it, we could probably find this, early stuff out for everyone.

But until now, there just wasn't, you would need your own clinician walking around with you 24 seven.

Sean: Yeah, Um, even, even in much more, Much less rather important aspects, right? Like it's the same way. I look at what we're doing here at log rocket, and it's the same thing where using AI to monitor digital experiences in real time.

Jeff: And we're able to flag to you something that's going wrong. You're an e commerce company, something's going wrong in your checkout flow. After it happens to five or six people, not after you've lost 300, 000. And yes, is that less important than helping someone stave off dementia? Yeah, I will say yes, absolutely.

It's less important than that, but right. It's all in the, where it fits in people's lives. 

Sean: think it's all important across all industries. I'm a big believer in AI for good. And to me if it's, A lot of it [00:28:00] is just anomaly detection. There's things that the human eye cannot detect in all this data, right? So whether it's looking at anomalies and security threats, or, flows in your app, that are being, or that go down or healthcare or on a manufacturing line. If you can detect it earlier, it makes everybody's lives better. Not just you're making more money because you reduce, churn and spend on, if a part breaks in your factory or your logistics, flow or someone's life, so I think it's really exciting to see all these companies starting to.

Build AI for good as well as AI that actually solves a business problem.

Jeff: Yeah. , I think the biggest thing we're going to see from AI is just The time to awareness, the time to answer, the time to knowledge is just going to decrease so much no matter where it's applied, which is a really cool, I think, just general thesis of it. And the other good news about Altoida is it actually works with pharmaceutical companies, the health companies and the [00:29:00] billing systems and all these lessons about go to market, channel fit seem to have lined up very well here.

Sean: Yeah, absolutely. And, with Altoida, we've been working with some amazing pharmaceutical clinical trials showing great evidence for them, to progress, certain patients into a certain drug if they fit or not and it's just also exponentially increasing, with protein folding and Understanding of what kind of combinations of drugs to build, based on that.

And then all the way through, I was mentioning treatment, there's it's transforming primary care primary care doctors, they're like a generalist and they have to be just good enough at everything to point you in the right direction. Now we can actually give you a POCUS, which is just, basically a little point of care ultrasound on your heart.

If they see any issues. And they can do a quick scan with that now to predict, within a pretty high accuracy. If you might have, heart failure with preserved [00:30:00] ejection fraction or something like that, where you can say, look, I don't know, I'm not a cardiologist, but Hey, I'm going to refer you now, because I saw something we would not, wouldn't have seen until three or four years, and then you could put them on the drug that now exists for that, that didn't exist a few years ago.

So it's really fascinating to see this starting to happen.

Jeff: I love that. I'm not a cardiologist but you should see a cardiologist and now we know this years in advance of where we would have historically known it. 

Sean: You know, It's like shunting off to the right. person, right? And that's the primary care doctor is going to do. Yeah, I think the primary care doctor is going to be that kind of central node to really help direct you. It's like your orchestrator to put you into the right path.

Jeff: Yeah. Now, I feel like I would be remiss if I didn't talk about your consultancy that you do, your consulting that you do, and your mentoring that you do also, because I think you picked one of the, my favorite names I've ever seen for a kind of, one person consulting company. It's [00:31:00] called the Shongularity.

Which I just love, but you mentor and you work with, and you consult with kind of startups around the Boston area. So I got to ask what are some cool use cases for AI that you've seen from these kind of up and comers?

Sean: Yeah I love mentoring young startups, coming through tech stars, myself it was such an amazing experience. It's where I learned, from some amazing people what product management even is, what agile methodology is, how do we do it? And it was like a MBA in four months.

And it was just so impactful in my life at that time. And so I love to give back, E for all, as well as tech stars. And this group that just graduated from Techstars Boston there I was mentoring this one startup called Vitel. Looking at women's health and they have this amazing data driven platform for paramenopause and detection and all that goes into, the changes in women's lives at that time in their life is [00:32:00] critical and to, and for so long women's health has not been really discussed and Now there's this flood of amazing new startups coming out that are addressing some really amazing applications like, you know, Paramount menopause, which I think we're going to see a lot of change in that.

So I'm really excited about what Roma and her team, she's a fantastic entrepreneur. So what they're going to do there's Praxis pro is one of my favorites. So they're doing AI powered sales training for . regulated industries like pharma there's radicates again, another healthcare one where they're creating synthetic pet images from CT scans, which is just bonkers.

Jeff: That's insane.

Sean: so there's all these amazing Very specific, industry solutions that are coming out. And there, people I think are less trying to build a 20 billion unicorn. And they're saying, look, we can make a really good business that could still generate a lot of [00:33:00] money by focusing on a real problem.

Again, we mentioned this earlier that solves, something that solves a real problem with AI.

Jeff: Yeah. And it's also great to see that not all AI companies are leading us towards Terminators and and, liquid robots, but rather let's solve healthcare problems sooner, faster, better. Let's make it, cheaper probably as well, because if you're doing it earlier, that's going to really help.

Sean, I don't want to take up your whole day here. I could talk about all this stuff. I think for probably hours between healthcare and we didn't touch on music, which we're

going to have to connect separately. But I do want to make sure I don't take your whole day. If people are looking for you, obviously I know you're on top tall for your consulting practice for the Sean Gilliarity.

So people should definitely check you out there. Are you on LinkedIn? Is that a good, if people want to ask you follow on questions. 

Sean: Yep. I'm at, I'm on LinkedIn, so just look me up.

Jeff: Awesome. Sean, it was fantastic having you on. This is super interesting. Love hearing , how one of their early hipsters of AI have progressed through their whole career here. And

Sean: I

might have to [00:34:00] put that on my LinkedIn.

Jeff: I appreciate coming on man. It was great chatting and let's definitely stay in touch. I want to talk about music later.

Sean: Absolutely. Happy 2025.

Jeff: You too.