Writing a Book with AI, One Passion Project

Tim O’Connor:

Well, Hi everybody. Welcome back to Kindrel Commons, where we believe AI is for everyone because we show you real people around the world who are using AI to make their life easier, smarter, and a little more fun. I’m sitting virtually next to my partner on this, Adam Fahl.

Hello there, Adam.

Adam Pah:

Hi Tim, how are you doing?

Tim O’Connor:

I’m doing great today. We have a neat guest that I’m looking forward to introducing you to. His name is Sophian Bensaou.

And he is in Kyoto, Japan. I’ve known Sophian for quite a while. He’s a really smart marketer, but we’re gonna talk about something that’s not marketing.

We’re talking about writing because he also is using an LLM to do interesting writing. So, Sophian, why don’t you introduce yourself first about who you are and then we’ll go into the topic.

Sophian Bensaou:

Sure. Thanks, Tim. And thanks, Adam.

I’m happy to be here. My name is Sophian. I work in the tech industry, have been for my entire career for the past 15, 20 years.

And so I have a background in science and technology, but I never really had a background in writing literature. I enjoyed reading books, but it was more of a pastime. And I discovered that people were writing books using AI.

And I thought this could be a good opportunity, not just for writing a book, but also to learn how to use AI because this was a technology that was coming up and everyone was talking about it, but I didn’t have an outlet. I didn’t have a project to work on and writing a book came about and I thought let’s dive into it.

Tim O’Connor:

Well, let’s indeed dive into it with your writing a book. So, Sophian, why don’t you tell us the type of book that you’re writing? I know you have a variety of ones that you’re interested in that you’ve been doing.

So kind of give everybody a little backstory about the writing, what you’ve been doing with it, why you’ve been doing it, before we get into the notion of how to use the LLM.

Sophian Bensaou:

Yeah, I’m working on my big project right now. It’s a nonfiction book and it’s on the Episode 1: Tom Piteria – A Business Executive In Ohio Using AI To Make Artisan Bread At Home diet and we’ll talk a lot more about it, I’m sure. And I’m also interested in fiction and so I’ve written four short stories that I’m looking to self publish as well, but really the main one that I’m focusing on right now is the carnivore diet book.

Tim O’Connor:

A lot of times people are working on something, like say writing before they go to the LLM. What was the spark that got you to decide to use an LLM to help you with this?

Sophian Bensaou:

So the main reason why I wanted to get into writing this book, even though I don’t have a background in literature was because I was looking for a side hustle, something that could make money on the side all while being something that I enjoyed doing. And so when I found out that self publishing was an interesting business model, then I started doing more research and I love watching YouTube videos. It’s just such a great resource.

And so I came about a bunch of people who write books using AI. And so when I saw that, it became a lot less daunting, the whole idea of writing a book. And so I just devoured all of these videos.

They literally have like tutorials, the tools that you can use, how you can plug them in with AI and all that stuff. And that appealed to me because I kind of come from a tech background. So that was easy to set up.

And so it’s not like I jumped into it not knowing how to go about things. I just really just followed tutorials online and that’s how I got into it.

Adam Pah:

Which is really, it’s funny, right? As you say, like as a tech person, that’s how it appealed to you to like learn about this. I mean, tech people aren’t universally well-known for scoring high on the English part of the SATs.

That’s kind of our weak point. But you’ve had this entire interaction, right? That was set up for you to learn this and engage, which is really great.

And kind of beats the old model, right? Of like going and taking a writing course at like a community college. I mean, did that occur to you as like a different way to engage or just totally supplanted by what’s now available?

That’s a really interesting question.

Sophian Bensaou:

I hadn’t thought about it that way, but you’re absolutely right. There’s no way that I would have gone to do a course to learn about writing or storytelling. And there’s a ton of them out there.

And now that I’m getting into it, I’m getting more interested about it. But you’re right that it’s the reconciliation between the tech aspect of it was this entirely new prospect, this new endeavor that I never would have been able to get into had it not been for AI and the joining of the tech with it. And so it kind of was serendipitous in some way, but at the same time, it’s like AI is having such a huge impact on so many different levels that it’s almost inevitable that it would work itself out that way.

One more thing, because I got into this maybe a year or two ago, I’m more of a text type of person. I don’t like working with video or audio like on a tech level of things. Like I like to play music, I like to watch videos, but I don’t really do video editing or audio editing, but text editing, that’s very easy for me.

And so when I came across this idea that I can write texts using AI for a larger project like writing a book, that really just made sense to me.

Tim O’Connor:

You know, adding to what Adam just said, why not read a book about writing a book? I mean, you go into Barnes & Noble, any bookstore, there’s gonna be shelves of books there about how to write a book. Why not that?

What do you think was different about using an LLM versus just read a book, to write a book?

Sophian Bensaou:

Yeah, well, it’s funny that you put it that way, because actually the way that I learned how to write a book was through watching YouTube. So actually it was from a video format online. And then after that, I discovered like, this is something that I want to get into.

And then I started looking into some other books that I can read about how to write, about what books are good for figuring out what are good techniques for writing, what’s good storytelling, all of that stuff. So a big part of writing as any author will tell you is reading. And so I had been wanting to spend a lot more time reading, but now I have the perfect outlet in the perfect project where it’s not forcing me, but it’s like, I want to read more, which I’ve already wanted to, but now it’s towards this larger goal of writing a book.

And so it all comes together quite nicely, really.

Adam Pah:

Anyway, this is actually a really interesting like diversion here. We’re really talking about how we all like to learn, right? That’s what it is or how you would describe it.

Like I always prefer to read. Like the last thing I’m good at is like watching YouTube videos, unless it’s a home improvement project where I’m like, oh no, what does the thing look like? Like I should really see someone tackle this live first.

But for you, you’re saying, I mean, do you generally like to learn via video, Sophian? Yeah, I think so.

Sophian Bensaou:

I think so because YouTube and video, it makes it so much easier to consume. I know a lot of people are all about social media and scrolling Instagram, but I don’t particularly care for those. What I like to scroll, I don’t scroll Facebook, I don’t scroll Instagram, I scroll YouTube.

And so I guess I’m curious and I have certain interests and I’m quite particular. I don’t want to get a lot of garbage into my own algorithm, into my own feed. So luckily it kind of works itself out.

And I typically have quite a lot of good content that I’m interested in that could veer me towards these avenues for learning. And so one of them was writing books. I’ve had many different interests.

And so typically how I get into them is just watching YouTube and scrolling. Even if I’m not watching the video, I’m just scrolling. I’m like, oh wow, I haven’t thought of that.

That could be interesting. And then I kind of go towards that. Like I wouldn’t jump into reading a book unless I had confirmation and validation from a YouTube video of somebody telling me like, oh, this is the way to do it.

Adam Pah:

Yeah, it makes a lot of sense. Like you like the longer form content and now you’re being drawn toward this longer form content of writing a nonfiction book.

Sophian Bensaou:

Absolutely, yeah. And I know many people are crazy about podcasts. I had my own phase with podcasts as well.

And I go back to them as well, but you’re right. It’s the long format that appeals to me.

Tim O’Connor:

So now you go through that phase, you get all that information that you had, which was wonderful. And now the moment of truth starts, if you will, when it comes to an LLM and you start working with that. Can you recall what that was like when you started to first work with AI LLM to start to write?

Like what did you start to ask it? How did you start to interact with it?

Sophian Bensaou:

Yeah, so I really just followed the tutorials on YouTube. I set up my entire workspace. I had the right tools and now I’m ready to jump in.

And the first thing you gotta do is you gotta brainstorm and figure out ideas for what you wanna work on. And then once you get that going, you gotta write an outline and all of that stuff. And to be honest, yeah, the first time, like when I asked, I copy pasted the prompts that this guy had on the tutorial and it just felt magical.

It was just like, wow, this is no joke. Like I thought AI was good, but I didn’t realize it was this good. And I remember back in the day, even Tim, you and I, we used to talk about, oh, we should use AI for this and that, but it’s not smart enough yet.

So you have to really push it. And so that’s kind of the mentality that I went into it originally, but it was already like far better than what I had anticipated, especially for this new area that I had no idea about. So if I had to describe it, I think it was a magical kind of feeling of, oh, wow, this actually works and it’s performed very well.

Adam Pah:

So you had it helped through the brainstorming, through the outline, but did you try to write the first draft of the first chapter on your own or did you have it sort of- Not even. Not even.

Sophian Bensaou:

Not even, yeah. And that was really the aha moment. Like the big first one, it was, I did the outline, and that already takes a lot of time.

In writing, there’s two different types of people. Some people like to create an outline and then work off of that. And other people, they like to write as they go.

But I’m more of an outliner. And so I basically had the structure, the skeleton for everything. And all I had to do is then just fill it out with prose.

And when you click and you say, this is the story beat. I want you to fill in the prose for this. It just wrote it.

And I was just looking at the letters, just forming on the screen. And I’m just like, what is this doing? Is this even doing what I think it’s doing?

And then when it’s done, I go back to the top and I read it and I’m just, I’m flabbergasted. I would never have been able to write that. But at the same time, it’s not, I was using Claude.

Chat GPT in particular has a lot of Chat GPT-isms. And with Claude, I found that it wrote very well. Like it could be close to the way that I wanted to write anyway.

So that was really eye-opening when it did that all on its own. And then from there, I just decided, let’s keep going. Like if it’s this easy.

And of course, every time I’m just rereading and everything, it’s almost like the story and the narrative is forming itself. And I’m the audience and I’m reading a book. Like it’s the first time I’m reading a book when really I’m actually writing it, which is kind of counterintuitive, but that was how it felt.

Like I wanted to see what would happen next.

Adam Pah:

Interesting, right? Just to say this, because it kind of transforms it into you’re the conductor instead of the musician.

Sophian Bensaou:

Yes, yeah, absolutely. And as the orchestrator of it, of course it comes with a lot of responsibility and you have to make it your own, but you’re not working on the minute details. You’re working at a higher level and seeing it work as it does as what you asked it to do, but you have to be there and confirm that this is what you want to do.

And as we’ll talk about, I’m sure it’s never just done. It’s never just copy paste. And that’s why in the world of writing books, there’s a lot of purists who say you’re taking away everything that makes writing so essential and so good.

You have to go through the actual process. Yes, that’s true. But if I’m getting started, this is one of the best ways to get into it.

And then as I work on it, I’ll learn as I go. But yeah, there is that orchestrator aspect to it. Absolutely.

Tim O’Connor:

How did you get it to have a tone that was a tone that you liked? Because there’s all kinds of books, they’re all written with a different tone. The author has how they want to say things.

To what degree did you have to train it or how did you train it so that the voice it was saying was the voice that you wanted? Absolutely.

Sophian Bensaou:

In those tutorials, he would actually explain how to give those instructions. And so I took those as a basis and I tweaked them and I basically started writing and all the magic was happening. But then every time that it spit out like a number of paragraphs for the story beat, I basically work story beat by story beat.

And so a story beat is basically this is what’s going to happen in the story. It can be one line, it could be two paragraphs, it could be seven paragraphs. And so I just work at that cadence.

And when it spits it out, I read it through and then I make changes along the way before I get to the next story beat. And this tool, the way that it’s done is that it looks back at what I’ve already written to guide the voice and the tone for what I’m gonna ask it to do next. So the more I write and the more I tweak it, the more it has a window context basically to see what style to emulate and then it’ll pick up on that.

And so the better it becomes at capturing my particular voice. And on a different note, because I was working on my nonfiction book, well, I guess fiction is a different story as well, but let’s start with a nonfiction. I have a style that’s very direct, but I don’t really have a style because it’s my first time.

So I’m just going with the flow and as long as it feels right to me, that’s okay. And that’s typically what’s happened with nonfiction. With fiction, I think I’ve had to make more changes myself, but as I write it, it has been smart enough to pick up on many different things.

A few times I had to go back into the instructions and say, oh, I want it to be written this way or that way. But in general, the more you write, the more it learns my particular voice, even though I hadn’t written anything yet.

Tim O’Connor:

People tend to have a challenge, Sophian, with conversing with an LLM. Obviously you’re doing a marvelous job with that. Tell us about that because you’re sitting next to a computer and you’re talking to it and that computer is kind of acting as if it’s an expert writer sitting right next to you.

So what was that like? Because it’s normally a uncomfortable new experience for people.

Sophian Bensaou:

That’s an interesting question indeed. I think I have a background in tech and so there’s a lot of frustration that comes with it because nothing really works the way that you want it to, but the key to it is persistence, right? If the first output isn’t great, you just keep working at it.

And that’s the same thing with AI. I keep asking for revisions. And if it comes back with something that’s mediocre, I don’t settle for it.

I’ll push for better quality. And really what I’m acting as is like an editor, as a curator. And eventually, essentially I decide what to keep and what needs to change.

And another aspect to it, I guess it’s the curiosity of it. I’m curious to see what the AI is gonna come back with. And so I explore it.

And it’s not like it has to deliver what I want it to. I’m just asking it and saying, do this for me. And then I see what it says and then if it doesn’t work, well, I’ll just keep trying.

And so when it comes down to the actual writing, I can be quite flexible. But right now for my non-fiction book, I’m on my third revision of it. And I’m going through the grueling aspect of it, which is the fact-checking.

And fact-checking is like, it’s crucial, but I need to verify the information. And when the AI starts hallucinating and creating scientific papers and referencing them and giving abstracts and summaries for papers that don’t exist, that can be very frustrating. And sometimes I won’t even catch it until like three paragraphs in, and I’m just realizing, hold on, this reference, this citation doesn’t exist.

It’s very frustrating, but you just gotta work with it.

Adam Pah:

I was just about to ask. I was like, with fiction, right? It could be anything, but with non-fiction, like you’ve chosen, the accuracy matters a lot.

And yeah, it’s really curious that you say right now. Yeah, it’s hallucinating. It’s coming up with these fake scientific papers.

And it’s one of these, I think it’s an interesting question, right? Do you think it would have been easier if you had taken the outline and gone and said like, okay, here’s all the source material I want you to pull from instead of letting it have free reign? Or do you think that would have just slowed it down for the entire project and kind of made you lose interest?

Sophian Bensaou:

Yeah, that’s really funny that you say that because I have been working on it just this morning and I came across someone who does a lot of writing that uses a lot of references. And they all say, you gotta work off of your citations first. Don’t write it.

And then you try to fit it with the citations. But I didn’t know. I just said, let’s just jump into it.

Because the fun part for me is just putting my thoughts into paper or on the screen. And now I’m doing the grueling work of making it fit. But I mean, I’m still enjoying it.

It takes a lot of time, but you’re right that it probably would have worked out better also working with AI had I had all of my citations and all that. That being said, the other thing I want to add to this is the way that I’m writing this carnivore book, it’s a health book, but all the other books that I’ve read about the carnivore diet, they’re all written by doctors. And so they have that authoritative voice and they can have that because they are doctors, but I’m not.

And so that’s also the key to how I want to write the book. I’m taking a different approach, but that also means that I really need to be very careful with all of the research that I do. And so I’m going along with it because this is how I did it.

And I didn’t know how to go about it beforehand, but it’s still working out. It just takes some time.

Tim O’Connor:

When you say you’re taking a different approach, can you elaborate on that? That sounds pretty interesting to me.

Sophian Bensaou:

Sure. I mean, I had my story in my head. It’s kind of autobiographical.

I’m talking about my journey throughout how I’ve tried and explored many different diets. And I want to take the approach where I’m more of a, not the authority on it, I’m not a doctor, but I’m more of like the Sherpa. Like I’ve gone through a lot of these different diets and I’m looking back and I’m seeing what I’ve achieved, what I’ve learned and all the mistakes that I’ve made.

And what I want to do is I want to look back and say, if you’re going through this, let me guide you. Let me help you up the mountain. And so that’s kind of the approach that I’m taking in terms of how I’m writing it.

And so when I wanted to write it, I wrote it because I had all of these ideas and I wanted to cover many different aspects of all these different diets. And so that’s what came up first. I hadn’t thought about reading papers and then writing a book around the papers.

That’s not the story that I want to tell. The story that I want to tell is my story. And then I want to supplement it with as much research as I can.

And in fact, what’s great about this entire process is that I’ve watched a lot of YouTube videos and I’ve read a lot of books about the carnivore diet, but now I’m actually getting to do the actual research and seeing what’s out there. Many doctors say many different things and they quote the scientific papers, but sometimes when I do my own research, I’m finding that it’s a little bit more nuanced. So it’s part of the journey and the discovery and the validation of everything that I’ve gone through over the 20 years of exploring different diets.

Adam Pah:

Which is, I really love that you went into that explanation because it’s, I mean, you’re writing nonfiction, but it’s really autobiographical in a way. Like you’re really having an AI help you tell your personal story that it doesn’t sound like would have ever gotten told in any other way besides you talking to your friends.

Tim O’Connor:

Right, yeah, absolutely. Let’s explore when you had to be the detective, right? Let’s explore that a little bit more because you do have to have that authority in there in terms of information that you’re gathering, et cetera.

Can you think back when you saw something from it and it said, hmm, that doesn’t seem right to me, or there was a concern in what, like how did that process work?

Sophian Bensaou:

Yeah, I think the best example is, it comes down to the fact check. That was really the most flagrant example because I kept writing and getting super hyped because it was giving me all of these citations that were supporting everything that I was saying. And as if you’ve spent enough time with AI, you’ll realize he’s a totally yes man, or if she’s a she, she’s a total yes woman.

They do everything to please you. So if you say, please find this and find the citations, they will hallucinate and come up with something that works and supports you, even if it’s not true. And so when I did this, it felt almost too good to be true.

I remember there was one particular thing that I was saying that was controversial and I was expecting to have a hard time finding the references. And it just gave me all these citations, almost like the perfect quote and everything. And so I really had to dig in and look for the paper and I just couldn’t even find it.

So ever since then, I learned my big lesson, which is that for a fact checking, you have to have an entire system. There’s dedicated software for that, that you can cross-reference all the different citations that you’ll have in your book. And for me, this is my first time doing this.

Maybe someone who took a class in history and wrote an essay and they’ve done that many, many times, but I had no idea. So I was just trusting the AI and it’s a grueling aspect of writing the book. So I was just like, okay, let’s see how far I can get, how fast I can get there.

And then it just didn’t work out. So in terms of detective work, I just kept pushing it. I just said, this is wrong.

This doesn’t exist. And then it’s almost like I’m talking to someone like an assistant or I’m talking to maybe, I don’t wanna say a child, but someone who’s like an intern maybe, or even a colleague and just saying, you got this wrong, you got this wrong. Make sure you don’t do this again and please try again.

And then if it still doesn’t get it, I keep asking, please try again, please try again. Okay, you’re getting closer, but you’re still not quite there. You’re still missing this and that and that and this.

Try again. And so I can imagine that for many people, that just sounds like too much work. Like forget it, just do it yourself.

But it’s almost like, I want to test it. I want to see how good the AI is. And I have confidence that it is good and it will get better.

But it’s almost like I’m working in that fashion where I’m pushing it and giving it positive reinforcement. I feel like I’m a big proponent of positive reinforcement. I’ve heard of many of my friends scolding their chat GPTs and being like, give it to me straight.

Don’t be a yes man. I prefer just pushing it in a positive fashion.

Adam Pah:

So you’re not in the beatings will continue until morale improves camp? That’s not my style.

Sophian Bensaou:

I hear that it has worked pretty well for other people. But I hope that if the AI robots take over that I’ll be further down on the list.

Adam Pah:

Yeah, that’s so interesting in your approach. I think the crux of it is still maintaining belief in your own autonomy where you get the quote that’s too good for something that should have been more difficult. And you’re like, this actually raises my spidey sense at this moment, right?

Like this is too perfect. So you’re still maintaining that understanding of like, this is an AI, right? It’s not perfect and it’s gonna create these errors or hallucinations that you’re gonna have to fact check.

But the reality is you’re spending a lot of time doing this detective work and running it down, right?

Sophian Bensaou:

You’re right. And the thing is that it’s actually quite deceiving because when you’re working with it and it’s doing a great job, you just lean into it. You just believe in it.

It’s just, okay, yeah, it’s got it. What could go wrong until it goes wrong? And then you start doubting yourself first before doubting it.

So it kind of plays mind tricks as well. So you gotta be careful. And of course, all of the AI suppliers and all of those services, they always say that AI can make mistakes and it’s true.

But then it also reminds me that they’re not technically mistakes. They’re hallucinations and people think that’s bad, but in fact, that’s not a bug. It’s a feature.

You want it to hallucinate so that it can expand its mind and go deep in different directions. If it didn’t have that, it would be too rigid and it wouldn’t come up with anything new. It wouldn’t understand, like when I’m asking it to do something a different way, it wouldn’t have that flexibility.

So it’s kind of a price that you have to pay for it being what it is.

Tim O’Connor:

That is such a fascinating comment that you just made there. Can you think about a time when it gave you something that you were like, wow, that is so interesting in this whole area of creativity that it offered something?

Sophian Bensaou:

Yeah, I have a perfect example. And it was while I was working on one of my science fiction short stories. And I just followed the routine.

I had my brainstorm, I had my outline, I had my chapters, I had my story beats and I’m just going one by one, one by one and it’s going well until it produces this story beat, the prose for it. And I’m reading it and I realized this is going a completely different direction. Like this is not at all what I wanted to say.

But then I took a minute to read it through again and I realized, sure, it kind of hallucinated but it was going in an unexpected direction that actually kind of could work. I started thinking more about it and I thought that that completely changes the entire book and my outline, but this is so much more gripping. Like this makes a story an actual story.

Like it’s the central concept for the entire book. And so that night I just couldn’t sleep. I just kept thinking about how would I make all of these pieces fit together?

And yeah, I went with it. I was grateful it hallucinated and it gave me that breakthrough idea that I would never have thought of. And I kind of ran with it.

Adam Pah:

I just love the context that that gives, right? And especially the context of what a hallucination or mistake is. You know, in a fiction book, it’s a real partner here.

In a nonfiction book, I mean, it’s annoying but I think to step back, you’re the editor, not the writer. And every editor has to deal with that with an unreliable narrator author that they have to fact check and it’s grueling work. And I guess I’m showing my age here, but I just, I always remember the million little pieces author and the Oprah interview where he was like, oh yeah, I guess it’s not entirely a memoir.

I may have made a few things up. And it’s like, that’s every editor’s nightmare is working with this author where it’s like embellishments in a nonfiction. And it’s like, oh, this is grueling.

But really, you know, it’s making mistakes but depending upon the context, they’re either happy little accidents or, I mean, if this was an actual human, you were working with something you would have had to deal with anyways.

Sophian Bensaou:

Yeah, yeah. And what’s interesting is when you, we like to think of nonfiction and fiction as being completely separate. And like we were talking about fact checking for nonfiction is crucial, but in fact, it’s just as crucial, maybe even more so for fiction.

If you’re writing science fiction and I’m not writing for, you know, like the high sci-fi, like Star Trek stuff, but those people need to have it grounded in actual science. If they say anything that is just a little off, like the fans would go crazy. It’s like, how dare you say that it’s two parsecs away?

Like I have a compendium right here that says it’s more than that, you know? Like it has, if you break apart those aspects of it, it breaks the whole foundation where you have that suspension of disbelief, right? So it’s almost more important to have that be well grounded in fiction than it is in nonfiction.

That being said, with fiction, you can go left and right in the directions within those parameters. And so where this wow moment kind of happened, it just happened to go off rails, but it was off rails, not in the foundations of it, but it was in my foundations, the ones that, well, technically with the AI we built when we wrote the outline, but somehow it just kind of said, yeah, I don’t know. Like I forgot about the outline, let’s go this way.

And it just kind of worked itself up.

Tim O’Connor:

Wow, that’s really interesting, Sophian, that experience. We’re coming close to our time, so let’s kind of start weaving into our closing. And what we’d like to ask is, if you were sitting down with somebody and they had some advice, they were asking you for some advice about what to do, what would you tell them what to do?

Like what would be like a first thing you’d say, go do this?

Sophian Bensaou:

I think, first of all, you need to find a project that interests you that you can use AI for. I did not have anything for a long time. I’m in tech, but I just didn’t really need it for writing code.

And I couldn’t think of any of the other interests that I had where I would be using it for. Luckily, I came across this entire idea of writing a book. And for me, that’s entirely new.

So even if it’s something that you’re interested in, but you’ve never dabbled into, but you’re seeing that there is a potential for trying it with AI, now is kind of the best time to give it a shot because AI is making it so much easier for you to jump into it. You don’t need to have a background in literature. You don’t need to have a background in whatsoever.

You can just go in and give it a shot. And I’ll add to that, chances are there are many other people who are facing the same issues. They’re trying to figure out how to do it.

Just go to YouTube, go find a book. There’s plenty of resources out there that people want to share how they’re using AI. And so that’s how I came about it.

And that’s how I’m working with it. And even to this day, I’m just finding new uses for AI. And it keeps surprising me.

And it opens up the realm of all the different interests that I can have now because I’m able to jump into these things because I have the help of AI and a companion, basically. Like it’s someone who can help me have a conversation about it.

Tim O’Connor:

Adam, any closing comments that you have you’d like to add?

Adam Pah:

I mean, I think Sophian just gave the best answer that’s possible. I guess I’ll just plug our own podcast here as one of those resources. But I will just say, I think the key there is yeah, finding your passion, right?

And then using it as a partner. And I think I’m just very much in my Bob Ross moment here, but accept the AI as that partner. And embrace the happy little accidents along the way.

I like that a lot.

Tim O’Connor:

Yeah, I like that a lot too. The happy little accidents. Because as you mentioned, passion.

And I think that’s one thing we’ve found through all of the interviews that we’ve been doing is everybody’s been doing something that is a passion project. And that’s so much important to everybody is that you start with something that is a passion project. Because when you get into this, you’re gonna have twists and turns and things aren’t gonna be clear.

And you have to be persistent. You have to be creative. If you know that you’re doing something that you have a lot of passion, you can stick with it a little bit more.

If it’s something that it’s not a passion thing, you hit a roadblock and you might just say, okay, it’s time to turn around and not go forward. So here’s another one of starting with something that is passion. As opposed to starting with something that is work.

And clearly people may have things that are work related that are passion. There’s no doubt about it. But the way to get into AI, the way to learn an LLM is to get something that really is at your heart that you wanna do so that you can kind of move through that and stick with it.

Because it’s so important that you get past the chasm, if you will, of the learning. Where you can get past just a little bit then you have the breakthroughs as Sophian has had. Sophian, I’m just so thrilled that you can join us today and share your knowledge with everybody.

And so thank you very much. Thanks for having me. And Adam, can’t do it without you, my friend.

Always good to have you here, tag team and you and I with our guests. So thank you very much.

Adam Pah:

Same to you. Thank you. Thank you, Tim.

Thank you, Sophian. This has been a really fun conversation.

Tim O’Connor:

Well, it has been a fun conversation, everybody. And again, at Kindrel Commons, our objective is make AI for everyone because it is for everyone. And here’s another example.

Here’s a fella in Japan doing this. We’ve had people all over the world using LLMs. It’s not just something for one type of person. It’s not, we’re just one type of project.

So thanks for joining us. And we look forward to having you again on Kindrel Commons. All the best.