Personality Driven Development: Exploring the Frontier of Agents with Attitude

Alright, everybody, my name is Ben, and I'm going to talk about anthropomorphized agents. I'm calling it personality-driven development, which is a kind of a cute name. And what we do at Perpetual is we build AI agents, we call them virtual teammates, or AI employees, and we have really leaned into the idea of giving them forms and form factors.

So, level set on terminology,right? Anthropomorphization, which I had to practice saying a whole lot. I like A18n, which I don't know if it will catch on. That is when you give human traits to non-human entities,right? So Yogi Bear, or Lightning McQueen, or Nemo,right?

These are all anthropomorphized creatures. Fun fact: zoomorphization is when you do the same thing for animals. Not really relevant to this talk, but I thought it was kind of cool. Aslan is a good example of this,right? Aslan is a lion from Narnia, so it was a non-human entity that— or, whatever, a non-animal entity that got animal characteristics.

You can tell your boss on Monday this is what you learned at the conference. So, these are not new ideas,right? We've seen this in software for decades,right? Clippy, you know, her— I actually put this on the slide before, you know, the recent OpenAI fiasco.

But anthropomorphizing, giving technology a human form, very, very common,right? And we're seeing this obviously a lot more with agents, but again, not new ideas. Right, so Mailchimp, you know, 20 years, they've had a monkey as the form of a human that, like, sends email for you,right?

So again, these are not new. But I would ask,right, so we have Siri, which is clearly not anthropomorphized,right? There's no persona here. However, I'm going to make you raise your hands: how many think Siri, or you could do Alexa 2, is female?

Anyone think it's male? Got one. Anyone think it's non-binary, or think this is just a stupid question? Right? It's very weird,right? And this is sort of a lot of, like, my learnings and my observations, is that we're all people, and we really like to ascribe human characteristics to things,right?

So, well, suddenly this AI became female because it had a voice. And that's strange,right? So, you know, GPT-4o, is this female? Male?

What about when you're talking to it? Suddenly it's like, oh, well, now it has a voice, and so suddenly it has a prescribed gender. And these are just weird concepts,right? And they're normal, there's nothing profound here. But it's interesting when you're thinking about product development and software development, what your customers or what an audience is going to perceive on the other side.

The foundational models, interestingly, have really moved away from any of these concepts, and you can guess some of the reasons. Some of them we'll talk about. But these are, like, the most modern representations of the foundational models, at least the ones that have consumer experiences,right?

You have Copilot, you have OpenAI, you have Meta AI, Gemini. This is how they're being represented,right? They clearly all share a single designer for some reason.

Actually, I want to go back to the female thing for a second. The— I'll keep it here. When we got a Google Home a couple years ago,right, the little screen you put in your kitchen that shows your photos, and I was super excited, and I was, you know, showing my wife, and I said, "Hey, look, we can set a timer, and it can play music, and all these things."

And she looked at me with just this, like, look in her eyes. She's like, "You cannot have a woman in the kitchen that you boss around." She's like— and she was dead serious. She was like, "That is unacceptable.

You can't just tell a woman what to do." And I was like, "It's not a woman,right?" And she was like, "I don't care." And she was very, very serious. And I was like, "Wow, this is remarkable, how, like, deeply ingrained this are."

So I changed the voice to, like, an Australian man, and like, now we're cool, and like, everyone's happy. But, like, true story,right? So, I think it's also helpful to think about the counterpositive, the opposite. Like, what is it, like, an AI that's, like, doesn't have a form,right?

So, this was just a great example,right? The AI that's inside Google Photos is just mind-blowing,right? Just, like, search for a dog, all the pictures of my beagle, like, lovely. But there's no form here. You don't think about Google Photos as having, like, personality,right?

It just— it just is. And the algorithm is, like, under the covers. Oh, another fun fact, as long as I'm talking to my family. So I showed this to my 11-year-old Zeke, and he was like, "Oh my God, Google Photos has ChatGPT inside."

So if you wonder how, you know, Gemini's branding is going with the youth, it's not—

not good at all. So, at Perpetual, we've really leaned into this idea of giving agents forms and personalities, and we really took it a couple steps. And we tell our customers, "You can give your agents their own forms."

And so here's Tech Lead,right, kind of a cyborg, android type of, you know, persona. And, you know, a member of the team writes code, does code reviews, things like that. But we really leaned in to say, "Listen, they can have a personality, they can have a form factor, they can have preferences."

And then things get real weird because our recruiter is, like, an artichoke. And it's like, "Oh, it's kind of like technology that's zoomorphized into a— or whatever you do with a vegetable that now has human characteristics," and it's all very bizarre.

But on the other hand, it, like, oh, it actually kind of makes sense. You're like, "I mean, it makes no sense, but it's also, like, I understand this." Like, it's a recruiter who has this form, and we have teams of agents.

And these are hamsters, and they run the business, and we have the general manager and the graphic designer, and they represent

their own roles,right? And on one hand, it's very amusing. So the question would be, like, "Why are we doing this?" And, well, I'll get to why we're doing it in a second. Let me talk about the expectations, because this was— this was surprising.

So, customer expectations, as soon as you put a form onto something, like, get real, and they get real very quickly. So, assumption number one is that you can chat with it. And there's nothing about workflow. I mean, we're— at the end of the day, like, our agents, all we're talking about is workflow, if we're being real,right?

It's just, like, it's just smart workflow. That's what we're all doing. There's no reason that you should be able to chat with it, but all of a sudden, "Oh, it has a face, I must be able to chat with it."

"Oh, do I talk with it in Slack or in Teams?" Like, wait, why do you think that should even be a thing? But, like, 100% it is. Personality. Everyone assumes it's going to have a personality,right? And generally, the baseline is, like, you're a helpful assistant,right?

So assume, "Oh, you're going to be friendly, a helpful assistant." And we've very, very quickly adopted that mental model by— by we, I mean, customers, like, have adopted this mental model. They're just like, "Oh, a helpful assistant. That's great."

You know, we let customers make, you know, make them snarky, make them funny, but, like, at the end of the day, there's an assumption. And again, also weird. Like, why does software have personality? And suddenly, with a face, it just needs it.

Users have no patience with these things going wrong,right? We all know this stuff goes off the rails, it gets wrong, but, like, all of software fails all the time,right? But you never hear people being like, "Fuck you, Google Sheets,"right?

But, like, they curse those hamsters. You better believe it,right? It's just, like, it's weird. You suddenly have this thing that you can get mad at, and, like, you can— and this is all just psychology. It's user psychology that is really, really innate.

And thinking about this from the perspective of product development, of, you know, software development, it's like, well, is there any reason that we actually do it? Like, it seems like there's— I mean, I already named some downsides, I'll actually list a lot more at the end, but so why do we— why do we even bother?

First off, these are really easy concepts to understand,right? When I tell someone, "Oh, you have an AI software engineer," okay, like, you instantly know what we're talking about. "Oh, it's an AI recruiter." "Oh, okay, well, it'll probably, like, read resumes, and it'll probably coordinate interviews."

And without any additional words, and we found that to just be an incredibly powerful— what's the word? Like, a jargon. Not even jargon. It's just, like, the terse way to describe the things that we're all doing without getting into really complicated discussions about React frameworks,right?

It's just great.

Branding,right? This is— I would say it's— I don't know if branding is quite theright word, but having a handle, like, something to describe what it is, is also really, really powerful,right? I think most people think of, like— actually, just talking about AI for a second.

Like, the algorithm,right? People, like, know the word algorithm, and everyone thinks an algorithm is just, like, why my news feed is not in order,right? Like, "Oh, it's the algorithm,"right? It's just this concept that is just very hard to grasp, very hard to grasp.

But giving something a name, or a name that's something that we are familiar with, really, really powerful, because now you can talk about it. And I was really struck by— I don't know for the Android folks— Google Now used to— all the phones are going to buzz.

Or I guess all the phones that haven't been updated in three years are going to buzz. Google Now was essentially what Google Assistant became, and it did all the same stuff. It set your timers, and it— your reminders, and you would talk to Google Now.

But, like, what was it? It was, like, it was just, like, this weird conceptual thing. But all of a sudden, it became an Google Assistant, and you're like, "Oh, I get it." It's a thing that, like, works on my behalf, and it, like, you have this, like, almost corporeal understanding of, like, what it is.

And I'd actually be curious if Gemini makes this better or worse. But again, Google Now and Assistant are the same thing, but just that one nuanced difference made it really easy to understand.

Price anchoring. So this is an interesting one. I don't know that this is, like, well-tested in the field yet. All of this is so new. But if what we're talking about, again, is just, like, workflow, and all of our agents are just doing, like, workflow, like, what is the mindset for what workflow should cost,right?

We look at comps, and it's like, "I don't know, 50 bucks a month." I mean, there's sort of, like, there's— again, making up the numbers, but if you think about it as a percentage, or if you're price anchored on a junior employee, wow, it changes the conversation,right?

So you talk to an executive, and it's like, "Yeah, 1/20th of the cost, 1/100th of the cost,"right? Suddenly it changes that nature of that pricing conversation. Again, I don't know that this is fully battle-tested or if it will withstand tests of time,right?

But, like,right now, it's fantastic. Is that a ferret? What is that?

An otter.

An otter? Hmm.

That's an awesome picture. So the other reason that this is a really helpful construct is it is a way to decompose problems,right? So thinking about if, you know, wearing an engineering hat for a moment,right? What is engineering? It is abstractions about the real world, it is getting theright levels of abstractions, and it is about decomposing problems.

Like, that's just all we do all day long,right? And in a sense, this is, like, an arbitrary way to break down a problem. On the other hand, it's a really useful way to break down a problem, which is what we've learned,right?

So specialized agents— I've heard a bunch of talks today talk about specialized agents versus generalized, and how the specialized ones perform better,right? It's like, "Oh, we have a finite set of tools. We have a small number of inputs."

Like, there's just less chance for LLMs who are trying to interpret or do tool calling to get things wrong. And so it just happens to be a really convenient way to break down problems, and also to scale, because we can keep subdividing, like, agent problems into more and more specializations.

And so it's just very natural. Again, arbitrary, but, like, even for me, it's just like, "Oh, this is very helpful. I understand that my AI engineer writes code, and I understand that my AI copywriter, you know, writes compelling copy."

And, like, great, I can get my head around that very, very easily.

And lastly, just fun,right? Which is, you know, that's sort of a company branding question, but, like, we like making our video game avatars,right? You spend more time making your, like, eyebrows correct in, like, the Nintendo Switch than you do playing the game.

We roll characters in D&D. Like, so, like, why not have fun,right? So that's sort of my personal perspective on this. Let's talk about the downsides. And there's not going to be any cute pictures because, you know, this is, like, the sad part.

So, okay, so one of the big glaring ones is, like, we were just inviting inclusivity and stereotype challenges,right? We were just asking for it. One thing that was fascinating is, you know, all those avatars were generated,right? So a customer can pick their form, and we auto-generate it.

100% of the software engineers are generated with, like, necties, like, looking like men. That's just, like, what comesright out of Dolly. And, like, I'm not going to have any, like, perspectives about, like, it's just, like, all of them do.

So we were just inviting this onto ourselves. And is that worth it,right? Is that worth it in a work context to really just invite those questions?

Expectations of performance. I don't know what it is because we assume, like, software is going to always just work, but there's this expectation that, like, these agents are going to perform really, really well,right? There's just this bar. It's like, "Oh, well, my, you know, even though our, you know, our junior employees don't perform well, there's an expectation that these things are going to perform at a very high bar."

It's just what we've seen. It's like, "Yes, of course it's going to get itright 100% of the time." The features I alluded to before, it's like, why are we spending our time building, like, chat interfaces and all of these things that, like, it's just, it's strange.

Like, you almost have to if you're building this type of personified agent, but it's just because it's expected.

Certainly, as a startup, we haven't had to deal with this, but I'm going to guess that when we want to walk this back and rebrand, like, holy shit,right? Like, there's just, like, walking back when all of our customers have these personas is going to be very, very difficult,right?

This is not just, like, changing some colors,right? This is a major, you know, stake in the ground that we would be planting. And lastly, it could be a distraction,right? Like, all of these fun stuff that we're talking about around personalities and, like, chatting and preferences, it's a distraction from the actual business value, which is, like, document review and data extraction.

Like, the thing that actually someone would pay for, it can be distracting. But at the end of the day, honestly, the biggest downsideright now is that it is just a very stark reminder that you're replacing jobs. And, like, the outside the scope of the talk, whether or not that is a good thing or a bad thing or an inevitable thing,right?

However, it is a reality that as soon as you bring this up with a prospect, like, first thing on their mind. And I'll just be real. I'll tell a story. So very recently, I was, you know, pitching sort of, you know, C-suite pitch,right, to a CEO.

I was like, "Oh, this is, you can't scale your business. You don't have enough people. You could never hire enough people to scale the business to meet your aspirations. Have we got a solution for you?" It's also shit work.

Like, no one wants to do it. Like, it's great. And he was just loving it. He's like, "Yeah, this is, like, exactly what I need. And at the cost, it'll be great." And so set up our first design session, and we get into the meeting, and he has, you know, one of his, you know, like, an IC on the team.

He's like, "Oh, yeah, well, I don't do any actual work. I'm, like, an executive,right? I brought the person who knows what they're doing. This is Ben, and he has software that has virtual employees. Can you please tell him what your job is?"

Like, my jaw dropped. I was like, "Oh, shit." Like, I was not at all prepared for that,right? Just like, "Oh, I can't talk to you with a straight—" like, whether or not I'm going to actually make you more productive and you will do better in your— like, just to lead with that in the conversation was just, like, really, really hard.

So I had to, like, very, very quickly, like, on the fly, try to, like, walk back that concept of, like, "Oh, no, no, no, no. Like, this is actually going to help you." And, like, whether— we don't know what the future is going to hold, but, like, we are definitely, like, leaning heavily into this, and it's potentially a huge hurdle.

Got a couple more minutes. I want to talk about, you know, sort of what was the title? Personality-driven design. And there's this— I don't actually even have, like, the words for this. I'd be curious afterwards if folks do.

The software we're building and our ability to, you know, create these roles, these virtual employees that have job descriptions and forms and personal preferences, and, like, there's no, like, checkboxes inside, like, the configuration. It's all very prompt-driven,right? But it's a way to inject nuance and business logic into

these agents with zero configuration, which means that every single instance is, like, 100%, excuse me, bespoke for each customer, which is a really wild concept. When you think about, like, "Oh, can I reproduce this bug? Does it work on my machine?"

It's like, "Well, of course not. Like, of course this doesn't work." Every single instance is bespoke,right? So here's just, like— and this is not a real one. This is an example,right? A giraffe,right, who has a personality. It's charismatic.

It's entertaining. You review resumes. You read cover letters. You know, you do that, like, operational work. And here's where it gets super interesting is the preferences,right? "Oh, as a hiring manager, I want to tell my recruiter my priorities.

And I like people who went to Ivy League schools, and I don't like job hoppers." I sound like such an old man. "And I don't like people who hop jobs, and, like, I like cover letters." Like, "Okay, well, I can train," you know, lowercase t, train my virtual employee the way I want to work and the way I want them to work.

And so this is just an amazing concept that, like, again, I really don't have my head around, like, what it means for, like, the future of software when an entire thing can be molded to meet the business's need, not just, like, "Oh, what the PM of the SaaS platform, like, happened to, like, think was a good checkbox,"right?

And so this is sort of, like, the future that I'm really, really excited to be working on. Period. Full stop. Awesome. Thank you, everybody.