← Liftoff with Keith: Conversations with Super Founders and Growth Experts sharing useful insights
Liftoff with Keith: Conversations with Super Founders and Growth Experts sharing useful insights
Agentify: How AI Agents Are Reshaping Companies, Careers, and the Future of Work | Michael Palmer
00:31:09
Loading summary
Transcript58 lines
- [00:00]
Keith
Foreign.
- [00:06]
Podcast Host / Announcer
This episode is brought to you by the outstanding team at Compass Strategic Advisors, your trusted partner in driving strategic growth. Whether it's expanding into new territories, launching products, rebranding, or hiring key staff and board members, they offer the expert guidance every step of the way. With a versatile approach and a proven track record, Compass is the go to resource for both startups and established companies looking to scale with confidence. Thanks and show your support for this podcast by sharing some sponsor appreciation. Check these guys out at bww.compass-strategic-advisors.com all.
- [00:38]
Keith
Right, and we are recording and lucky enough to grab the author of a brand new book called the Art, Science and Engineering of Successful AI Agents. And here we are with Michael Palmer, the CEO and chief scientist of Taos Research Corporation and the author of the book, Michael, welcome to Liftoff.
- [00:59]
Michael Palmer
Keith, thank you for having me. It's great to be here. I've been catching up on your episodes and you got some great shows.
- [01:05]
Keith
I like them. We have had a fun time. And what a fun year it's been of 2025. It has to be the year of AI. Some people might call it maybe the second year or third year, fourth year, depends when you got started in all of this. But. But it's become such a cultural phenomenon now. I'm also kind of curious, obviously, to dig into the book, but also to talk about what we should expect in the new year of 2026. Will that still be the year of AI? And what is the new layer of focus as everything from corporations that are focused on digging deeper into AI to find better quality output, better performance, better efficiencies and productivity. And then we are going to spend some time talking about how AI companies, those that are born of the AI era, how they should be looking at building their businesses. Is it one or two people? Is it 20 people? What are you doing about. About a building product, taking it to market, raising funds and doing all that stuff. And I've got the perfect guy because Michael. Well, I'll let you tell your background, but it's a great combination of all of those factors.
- [02:18]
Michael Palmer
Well, thank you. Yeah, no, I think you're right. And it's a great framing that this is. Things have been moving at such a fast pace, honestly, it's hard to keep up with all of the trends, all the things are going on. And my background is I've spent basically my whole career as a builder. I'm in the east coast now, but I was in Silicon Valley for most of my career. Started in robotics and moved through a Series of big tech startups, small companies, and then ultimately I wound up as a CTO at a large bank for about the last seven years or so before founding Taos. And I was doing AI there and what have you. And I stayed through sort of maybe the first year and a half after the GPT moment happened in late 2022. And I just realized that, like, I had to get out again and start working with smaller companies and some bigger companies. But the rate at which things were happening, I knew I just couldn't sort of sit inside a big company and. And be in touch with the pace at which things were happening. So that was my kind of motivation. And it's been a whirlwind, you know, I think for everybody in the space, the change. I'm a coding developer. I still write a lot of code. And my workflow for building things literally changes every. At least every month, if not every two to three weeks with just the new models coming and the new tools and the agents and what have you. Are you.
- [03:45]
Keith
Are you a lovable fan, a Vibe coding fan? Are you shrinking that and compressing that time to develop?
- [03:51]
Michael Palmer
Very, very much so, yeah. And I'm kind of in the. I actually have a chapter in the book called Beyond Vibes.
- [03:57]
Keith
Yeah.
- [03:57]
Michael Palmer
Rigorous AI software. AI assisted software development. And I think Vibe coding was such a great term when it came out. It captured the imagination is what we were all doing. Yeah, we're all kind of like learning how to use these different tools, whether it was lovable or replit or bolt or CLAUDE code or what have you. And now I think we're entering an era where people are realizing, wait a second, like, if I'm going to do this right, software still has nuances and bugs and security issues and what have you. Just because I'm able to get out a prototype quickly doesn't mean that it's high quality. And so it's kind of. I think that we see that shift underway now. And I think the same thing applies to people who are trying to build for the AI market or building applications or agents or other things. It's. People have been building a lot of stuff and then they find that the big AI companies just come and steamroll over them and build the same feature and like what you thought you were building. Yeah, OpenAI just announced that. And so you better be back to the drawing board pretty quickly.
- [05:01]
Keith
It is funny. You are right. This fast Revolutions. Talk to me about the motivation. The catalyst, it said, okay, I gotta write a book now because it is moving so Fast. You feel like if I write a book, it's going to be outdated in six months. How do you cover things like art, science and engineering of successful AI agents? And then by the time the ink dries, hypothetically speaking, you know what. Yeah. What happens?
- [05:29]
Michael Palmer
Yeah, no, that was, that was one of the big challenges I wanted to figure out. How can I sort of take a step back and think about the things that are more lasting? What are the things that are gonna be valuable to founders and valuable to maybe investors and builders of all sorts, you know, even as new models keep dropping and new tools keep dropping and what have you. So a lot of the effort was figuring out, okay, what are the things that are changing very radically. And I think we have a lot of radical changes underway. I think a lot of traditional software that we see is, is going to be replaced by a more service oriented agent that does more of the job for the consumer or the enterprise. And so less of this kind of clicking buttons. We spend a lot of time building these things with buttons and knobs and what have you. Now the agent's going to take over and do those tasks for us. The big thing I think people can do is focus on really getting into specialization, really know your customer, really, really get deep into what are the problems you're solving. And then the next piece, and I talk about this in the book, is adaptability. A lot of the agents that we see today, I think are going to be eclipsed in the next one year to two years, very, very much by agents that actually take the initiative. Right now, everything that you do, you have to prompt it. You kind of say, okay, Claude code, let me sit down here, I'll tell you what to do. And it'll go off and it'll complete a task where you maybe do this deep research or, you know, if it's a lovable or something, it'll build your prototype, but it doesn't then kind of keep going and understand your objectives. It doesn't actually kind of have its own goals and its own ability to take these. That's actually just around the corner. And there, there, there are a lot of engineering teams who are working on kind of that next level of, of agent that can take the initiative, actually function much more like an employee and, and pick up the ball without being told what to do.
- [07:26]
Keith
So Michael, on the surface, I'm thinking that that sounds more like AGI or some type of original intelligence behind the agent. Is that, is that what you're referring to or am I, is it just some more Elevated activity that is agent based.
- [07:44]
Michael Palmer
I think it's, I think that's a great question. That's a phenomenal question. And that's actually at the heart of one of the things I try to get at in the book. I basically try to say, look, in my opinion, we're not, with the current LLM technologies, we're not going to get to AGI. And I get into quite a lot of details to why. I just don't think it's going to be, it's not going to replace some of the type of thinking that we can do. But a lot of problems are being solved, there's a lot of scaffolding, better tooling, this new area called context engineering, where you really get very task specific and very domain specific and you give the LLM, you know, solve the short problem and they can do it, they do it better and better. And so effectively, I think what we're going to see and then we're going to simulate, we're going to simulate this more agent, true agency, true initiative. Even without AGI, I think we get maybe 80, 85% of the way there and then there will be people who are still working on more. That next breakthrough of true intelligence that.
- [08:40]
Keith
Comes outside of a prompt that comes out of some other type of.
- [08:45]
Michael Palmer
Yeah, it may come out of a, yeah, exactly. May come out of a totally different type of, you know, possibly systems where they prompt themselves internally. If we, if we kind of think of our own imagination or our own motivations. If you think of your, how you pursue goals as a human in life, it's some form of internal prompting, right? Like it's, you're, you know, you get up in the morning, you think, what am I going to do today? You know, what am I going to ask my agent?
- [09:08]
Keith
What should I do today?
- [09:09]
Michael Palmer
Yeah, yeah.
- [09:10]
Keith
So what do you call the most misplaced idea people have about AI agents today? Have we?
- [09:17]
Michael Palmer
I, I, I think it's, it's that what we're seeing today, agents, what I call agents without agency, agents that just respond to a prompt that that's really what an agent is going to be. And so I think most people think they see what's here and now and they're not, they're not thinking of, of the change that's really already underway. But it's going to show up in, in the next, if not one year, then certainly in the next two years where we see much more agency and autonomy from these things.
- [09:44]
Keith
And I like your point early on about how that's going to be A lot more focused and narrowed to the specific task at hand. But how does that get implemented at a company? That's one of my, one of my big questions today. It feels like there's a lot happening within organizations to, to leverage AI, but how is it being orchestrated in such a way that you know, groups are talking to groups and people are communicating with and, and there's a, there's a, there's a efficient collaboration going on versus more silos and more individuality in terms of, of process and activity.
- [10:23]
Michael Palmer
I think that's a super important point. Yeah, I think it's a super important point. So with the big companies that I work with, I hear a very, very common story which is we're adopting AI. We're, we're doing it. You know, they're, they're under a lot of pressure. All the managers of various businesses are under pressure because the CEO is under pressure. Everyone is telling them you've got to be doing AI faster, you've got to turn into a more AI first organization, whatever that means. And so they're, in practical terms they're adopting kind of point solutions in different parts of the company. Often it's, it's the call center. You know, maybe they're replacing those terrible chat bots that they've had for a lot of time.
- [11:03]
Keith
Yeah.
- [11:04]
Michael Palmer
Or a lot of folks are of course doing stuff in the development organization if they've got software developers in marketing. So there's a lot of kind of point wise adoption but not yet much sort of holistic understanding of the enterprise and more of like that over overarching kind of transformation. And I think people are trying to kind of supplement that with just educating their workforces, getting people to use more tools. Just, you know, agents are popping up in your, in your, in your Microsoft Word or your Google, Google Docs and, and encouraging those people, employees in the company to just use those tools and keep investing in as much they can. I do think also stuff will be carried in from the consumer side like we've seen in so many past technology revolutions where the consumers latch onto something that they just, they get addicted, they bring that into the enterprise.
- [11:55]
Keith
What become the metrics or the milestones for organizations to say, oh they're really using AI effectively. I would say, well, there's an SGA number, there's a hiring, you know, how much, what have we done with the workforce? There's probably a whole bunch of different ways to sort of track that. What are you seeing in terms of, of mid to large companies, how are they Measuring their AI implementation and how well they're doing with that. Right.
- [12:24]
Michael Palmer
I would say the, the honest answers, it's. First of all, it's a great question. And I think the honest answer, I think it's kind of all over the map. I mean, you see people who are, I hate to say it, but kind of frozen in the track, sort of waiting to see what the group, they're waiting to see what people are doing, kind of hoping to get a clearer message. Other people who are adopting in various places, but like, I think people are using things like as simple as, like how many people are using the internal, the internal LLM that we gave them and just pure simple adoption numbers which are not very good. They're not much of a thing. You don't know, right? You don't know. Are they using it for anything that matters to the organization? It's gotten, I think, a little bit better in software development just because I think that that is the most, the farthest along department, if you will, or function where. And then call. Call center where they're, they're measuring. Okay, is this, is this handling more calls or are we pushing out more software per, per unit developer and, and reducing. And can it reduce costs for us? But I think we're in the early stages of truly judging effectiveness in a more holistic way.
- [13:34]
Keith
I'm really curious to watch that. You know what else I'm curious about? Let me throw you a little hot potato too, Michael, is the idea that there's ServiceNow, Salesforce, Oracle, Microsoft, Google, I mean they're all leaders in AI today, but I also see where all of these AI first companies that were born in the AI era starting and doing some amazing things. And I just wonder who's going to build that, that successful AI agent that gets adopted by a company and becomes the standard and pushes out maybe some of those older standards or forces one of those older companies to buy them or stuff. How do you see this evolving? Because I know we're gone, you know, SAS is still a tremendous, you know, platform, but obviously it's transitioning into, you know, some, some combination of AI SaaS or whatever the adoption models become from a, from a business model standpoint. But, you know, how do you see that AI first companies taking on the behemoths?
- [14:36]
Michael Palmer
No, I think it's a real, real, real struggle. And I think there are, I've talked to a lot of smaller founders and startups that are trying to attack those spaces and oftentimes like, I'm mostly giving them advice as a Former cto, someone who sort of sat in the purchasing chair. And, and I think it is a lot of younger founders particularly and newer companies. They don't realize just how much of a, how much of a hold, I don't want to say stranglehold, but a hold that the legacy vendors do have on these enterprises. It's very hard for the enterprises to even know what software they're using, where. And they've been very, they're very, very deeply penetrated by the salesforces, the Adobes and all these other tableaus and analytics tools galore. Right. And again, it's very siloed. Different departments will use different things. So what, what I, what I believe and what I write about in the book is that I think teams, teams that want to start companies now and be successful have to be quite, quite, quite laser like vertical, focused on a specific problem in specific areas and show they can do that. Well, the other problem they're facing is that like if you just go in and say I'm going to build a generic agent for the enterprise, well then you're competing with Anthropic and Google and Microsoft and Oracle. So your problem's not much easier in that regard. So I think the ones who I'm seeing having success, they're really going after concrete problems and concrete industries that are not, that are not the domain of just your office software. So there could be things in fraud detection, for example, or things in specific tools and quality control or agentic selling. I think go to market is another big one where there's room for really interesting innovation as well.
- [16:20]
Keith
I think so too. We talk about it at the GTM firm that I'm a part of where it really is go to market part of an ecosystem as opposed to going out on your own. Even if you had an AI SDR kind of a model or some other sort of agent driven focus. It's sort of like, well, it sure would be nice if you had a hyperscaler relationship or an LLM partnership or you know, combination of those and, and going to market with some of those bigger players or other, other folks that you can provide a full solution.
- [16:52]
Michael Palmer
Yeah, and actually that's the key word, I think you used the key word which is that enterprises for the most part don't want to buy building blocks that they've got to assemble together. And I think that's hard for sometimes us in the builder side to understand. We love to fall in love with our technology, talk about how it's built and what's under the hood. And what have you. But that's in the enterprise. There's just so much pressure to solve the problem that the integrated, the thing that comes with a full solution tends to win. And yeah, I've seen, I've seen some smart startups that are going to market maybe under the wing of a Microsoft or under the wing of a Google or someone like that where they're plugging into the ecosystems that they've already got established. I think those are smart things to do because you're. It doesn't mean that they won't potentially replace you. If your feature looks good next year they might say oh, we can add that too. But at least you're not trying to build all this stuff that I think.
- [17:50]
Keith
We'Re going to see a lot of that in 2026. And on that topic, another hot potato. What's going to be the one or two things that, that surprise us in this new year you're studying this AI world, Is it going to be the, the rise of the start of the vertical agent, the. Well, I mean, or what one, one development are we going to see from one company?
- [18:11]
Michael Palmer
Yeah, well I think there's one thing that I just view as an inevitability which is we're going to, I don't know if it happens for sure in 2026, but it'll happen like I would say for sure in the next three years. I think we're going to see the Solopreneur billion unicorn. I think the first kind of literally person running their operation with a lot of agents. And you also see this, you see how AI native born companies, as you said, they're not just adopting it, they're not just building it into their product but they're adopting it in their go to market and everything, everything they do their support, their service, everything else and it allows them to save a lot on capital. You know, they don't need as much capital as they do now. They're maybe have a pretty hefty token bill that they're paying to the model providers but still it's a lot more capital efficient than building a 30, 40 person team. So I think that kind of ultra lean startup is now in the imagination. I think we're going to see some examples that really shock people as to how small the company is. Whether we see that true solopreneur billion dollar or not, I don't know.
- [19:16]
Keith
But I, no, I like that forecast. Obviously there's been some talk about that and we're not quite there yet. But the breadcrumb Trail has started.
- [19:24]
Michael Palmer
Yeah, yeah. And then the other thing I think is, I think we'll see some cases where probably not in highly regulated industries or very risk sensitive industries, but I think we'll see some places where people really do give agents kind of almost full autonomy. Not completely let them do anything they want to within guardrails and within other things will essentially let them function more or less as complete, as complete employees. And in many cases those will be literally in lieu of, in lieu of hiring a human for the same role because it's taking on a much more holistic sense of the job, using different tools and systems, calling APIs, responding to the customer and doing all those things kind of in the way that, that a human would in that function, but and taking its own initiative to keep improving.
- [20:16]
Keith
So, so go one more step with me, Michael, and tell me the, the critical elements of an AI agent. Then in 2627, what do you have to have if you're building a new company?
- [20:28]
Michael Palmer
Yeah, yeah. So, well, I would, I would say my book is a pretty good start. So Identify, which is on Amazon, dives into exactly that topic kind of from the, what's the strategic product angle, but then quite deep into the engineering of what does it take to have an agent manage its own goals, act on its own goals and be able to kind of continuously plan and adapt to an environment? One of the things that a lot of agents don't do is that they don't actually observe the environment around them. They're pretty much their observation as a prompt. So you have to build in kind of these loops or where the agent is not just waiting for human input, but is actually observing. It could be observing your, your server fleets, for example, your website, or it could be waiting on inboxes and things of that nature. But it has to be observing and then taking multiple forms of actions. And you have to have that continuous process of observing the environment, planning what to do about what you see in it, and then taking action. It doesn't mean you don't ever check in with a human, but I think it's that, that continuous operation that's, that's key. And yeah, I get into quite a lot of details of sort of how to do that in the book and.
- [21:38]
Keith
The, the detail, and I should point out your background includes some time with Kleiner Perkins, the venerable VC firm on Sandhill Road, and you know, obviously making huge investments, you know, going back to like Apple Computer. But you spent some time there and, and if I'm not mistaken, I think there's a chapter in the book about building, you know, these new AI companies. And I think you guaranteed like 10 million in venture funding from Kleiner Perkins if they followed your instructions to the T. Oh, no, that's right.
- [22:07]
Michael Palmer
Yeah. All the instructions I can pretty much get. Yeah. Your funding is. Is in the bag.
- [22:12]
Keith
It's almost guaranteed.
- [22:13]
Michael Palmer
Almost. Yeah.
- [22:15]
Keith
Well, what, you know, another hot potato. This is my hot potato interview, I guess. What, what are you gonna build? Like, what if you were to start a company? We'll put it in the hypothetical. I don't want to go into your lab today, but yeah, we're going into Taos. I love Tahoe. Taos, Tampa. We're.
- [22:31]
Michael Palmer
We're New Mexico, so it's. I grew up in New Mexico, so it's a reference.
- [22:34]
Keith
I love it, man. That's the most beautiful place. Anyway, what are you building for? For if you're starting that company? I'm guessing a vertical on the agent side, but maybe not.
- [22:45]
Michael Palmer
No. Yeah, I think it is vertical on the agent side. What I've learned is that vertical is. Sometimes we say vertical, people think, oh, it's got to be very. A single industry specific. But sometimes vertical can mean sort of very, very functionally specific, sort of deeper in a particular function than other people have cared to go. But, but to your point, one of the areas that I'm. I'm quite interested to have my eye on is this problem of coordination that you mentioned. This sort of like, well, I'm adopting this point, solution, this point, this, this here. Does that mean I'm an AI first company now? What's the bar? How do you get kind of the whole company to be sort of moving forward with this? And quite frankly, how do you, as a leadership team or maybe as a management team, how do you identify where maybe you've got pockets of resistance, the middle management that maybe doesn't want to do this, or people who are pushing back on this. And you know, you've got to achieve a strategic plan. How can AI almost help with that whole company alignment? Just an example. Maybe almost everyone can relate to who's. You probably have done your goals. You do your goals with your manager at the beginning of the year. And in some companies, if you're doing a pretty good job, you check in maybe once a quarter and say, how are these goals progressing? But they're not terribly dynamic. They're not really responding to the day by day, week by week flux of what's going on in the marketplace or in your job. And so you get to the End of the year and you kind of take all your work and you say, look, I met my goals, but I do think there's a place for much, much more alignment between, you know, what the company is trying to do and like what are the different layers of management doing? Are they pushing?
- [24:25]
Keith
That's more, that's more prescriptive and metrics driven.
- [24:29]
Michael Palmer
Metrics driven, data driven and sort of aware of where the bottlenecks are. So I think there's this opportunity for agents to help find the bottlenecks, see where work is maybe not getting done or work is not aligned to the goals and then help not just prescribe but help the managers, help the leaders. Because oftentimes, as you said, it's silos. They don't, sometimes the people are operating in silos and they don't realize how out of alignment.
- [24:54]
Keith
So am I seeing it some sort of like a performance index that I can check in on and then it could say here's how well you're doing and here's where you're lacking and here's maybe some recommendations or things you should start to enlist. I like that idea.
- [25:08]
Michael Palmer
Exactly. And something that could operate not just at the CEO level or the, or the C suite, but they could operate down the organization and help people understand what's going on in my team, you know, what's going on in my, my organization.
- [25:20]
Keith
It almost strikes me that there's a sort of a new organizational model developing in this type of company because I don't need you in a role as I'm so, I'm so outcome and performance based. I just need a place for us to be able to input how well you're doing on your performance and that we're all aligned. So that's, that's almost like the AI coordinator role in a new company, which I don't know where that would fit in a traditional org model.
- [25:48]
Michael Palmer
No, I think, I think people are struggling with that and I, I don't have a great answer as to where it like ultimately should but you know, a lot of, of us adopted Agile, you know, practices over the years, brought in scrum masters and Agile coaches and you know, we did all the ceremonies and really a lot of improvements, you know, to how we develop things as a result of that. But now when you look at those processes, you know, two week sprints and you know, the grooming tickets and these sorts of things, it's kind of out of date with how fast things are moving. And I think we need something that operates much, much more at the Speed at which the AIs are able to move. And I think that changes the roles. So product managers nowadays can create prototypes and talk to people in the market and say with real functioning software that may not be perfect under the hood, but they can really work with customers and salespeople to say, is this what you want? And then go back to the engineering team, say, yeah, we know this is what they want. So instead of writing these long PRDs like we used to write in the past, or having months and months of meeting your cycle times, all of these cycle times just become much, much faster because every role is empowered to do stuff that they could never do before. So the role definitions kind of have to change.
- [27:04]
Keith
Yeah, because I think you're right in that it's faster and better, but you still rely on that interconnected flow. And I'm not seeing where those places are connecting. Although I'm not at a big company on a regular basis, I work with mostly early stage companies that still operate like early stage companies have always operated, although they use these tools at their discretion to improve their output and their performance. And I think they're gonna lean in more and more and be. Be required to show that level of performance to stay in their job.
- [27:39]
Michael Palmer
Yeah, I completely agree. I think that's. I think the smaller companies and the ones that do it at a not met, not an enterprise sale, they will quickly show that level of efficiency and then the enterprise will probably either have to buy them or get disrupted by. By them.
- [27:56]
Keith
Well, this is partly the science and engineering of it all. When you say the art from your book, again, I have to steal a chapter. But where's the art part of this coming in? Where do you see it from that? And then I'll let you go. I know we're running out of time, but I think from my standpoint, as, as somewhat of a creator also, you know, and somebody who deep appreciation. I'm. I'm always curious of where you'll see AI start and stop.
- [28:24]
Michael Palmer
Yeah, yeah, no, I. Well, I think one of the things. And I actually have a chapter on this in the book. I talk about product managers, behavioral psychologist. And this was. There was actually a job listing that OpenAI had that told me, oh my gosh, I've got to dig into this. You're starting to see very, very serious needs for product managers who can, who can work on the personality of the agent and the style and the way in which it interacts with people. So no longer are you just. You'll be building a lot less sort of traditional UI where it's buttons and stuff for people to click the traditional graphical design. But your design now shifts to what's the personality of this, of this agent that I'm interacting with? How does it differentiate itself? And we can already see know differentiation between like a, a Claude AI and an open AI and a Gemini. They all act a little bit differently and people are responding. Then I think that's an opportunity for startups to think about that whole interaction model. Like, you know, you don't, you don't want. There's this term sycophancy in AI too where you, I mean it's the, you're absolutely right that you get back from the, from the models and you're like, oh, just telling me what I want to hear. I think that's an opportunity for founders and designers to really think about what builds trust. How do you make your agent feel like a really trusted partner? And that means it has to behave much more nuanced way than I think what we're seeing. So I think that whole behavioral side is a, is a design, design space that's very new and open and hopefully fun for people.
- [30:04]
Keith
That is excellent. Well, well again the book is agentify the art, science and engineering of successful AI agents. It'll be interesting. I can almost see a, a sequel coming in a year or two listing it all the new things that have, have progressed in that time in a. A new set of forecasts as well. But guys, check out Michael Substack at Taos Research and Taos Research Corp if you want to get a hold of Michael Palmer. Thank you sir, for joining us. What a fun time. And yeah, I think your book is just, is just so well timed and so on point. I was glad you could share a little bit with us.
- [30:45]
Michael Palmer
Okay, thank you for having me on. A real pleasure. And yeah, I appreciate you having me on and look forward to seeing what happens in the year ahead as well along with everyone else.
- [30:55]
Keith
No, you're in the catbird seat. I'm jealous. So enjoy the success of your authorship at a lovely holiday season as well.
- [31:05]
Michael Palmer
Likewise. And you too.