%20(150%20x%20150%20px)%20(2).png)
Invest with AI
Brett Caughran and Khe Hy lead a deep dive into AI for investing, joined by guests at the cutting edge of the field. Our goal is to be your Sherpa through a rapidly changing landscape by distilling what's working, what isn't working, and how you can leverage AI in your own process. Follow along as we tackle AI's biggest challenges and opportunities, one episode at a time.
Invest with AI
Stoic Point Co-Founder: AI is Bringing Back the Lean Hedge Fund
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
In this episode, Brett Caughran and Khe Hy sit down with Raj Shah, co-founder of Stoic Point and a former partner at Light Street, to get into how a two-person fund can run resourced like a firm ten times its size. Raj makes the case that AI lets a lean, concentrated fund compete with much larger teams and argues he's more worried about AI replacing him than the junior analysts everyone else is fretting over.
We get into:
- How a lean fund recreates the institutional resource stack without the institutional headcount
- The three buckets where AI fits the process: screening, research, and monitoring
- The "meta-screen" that surfaces ideas across 50 filters at once
- The black-box quirk where the same prompt run twice returns two different stock lists
- Separating the deterministic screen from the non-deterministic one — and why he starts in Bloomberg
- How automated monitoring caught a read on Lux Experience from an unexpected place
- Why AI makes a strong junior analyst more valuable, not less
- Turning your own process into an intern training guide — and a sparring partner that rips apart a pitch
- Why Excel with AI was the biggest positive surprise of everything he tested
- Single managers vs. platforms, and what an AI-native fund means for raising capital
We're not coming at this as "experts" with all the answers. We're in it every day, testing, breaking things, and trying to understand where this is going. The goal of the podcast is simple: bring you along as we learn, and give you a clearer view of how AI is actually being used in investing. If you work in equity research, at a hedge fund, or on the buyside and you're trying to make sense of AI, this is a good place to start.
***DISCLAIMER: Everything you hear on this podcast is for informational and educational purposes only and should not be considered investment advice. Any companies, securities, or strategies mentioned by our guests or hosts are discussed for illustration and shouldn't be taken advice to buy or sell. Markets carry risk and individual situations differ, so please do your own research or consult a licensed financial advisor before making any investment decisions. The views expressed are those of the individual speakers and don't necessarily reflect those of Fundamental Edge or its affiliates.
Chapters (Timestamps)
Timestamps:
[00:00] Intro
[01:21] — Greenhill to Highline to Light Street: Building Stoic Point
[04:00] — Recreating the $5B Resource Stack at a Lean Fund
[06:30] — The Three Buckets: Screening, Research, Monitoring
[12:00] — Same Prompt, Two Different Stock Lists
[14:39] — Deterministic vs. Non-Deterministic Screening
[15:31] — The UI Problem and the "Meta-Screen"
[17:54] — When Computer Use Got Good Enough to Click Through Bloomberg
[18:49] — Can Codex Run Your Screens Today?
[20:00] — Automated Monitoring: How AlphaSense Caught the Lux Experience Read
[22:30] — The Narrative Violation: Why Juniors Get More Valuable
[25:16] — Turn Your Process Into an Intern Training Guide
[26:55] — Building a Sparring Partner That Rips Apart a Pitch
[27:36] — Excel + AI: The Biggest Positive Surprise
[32:35] — If Big Firms Automate Juniors, Does the Pipeline Break?
[34:41] — What Happens When LLMs Develop Judgment
[35:59] — Measuring ROI in P&L, Not Hours
[38:25] — Single Managers vs. Platforms in an AI World
[43:42] — The Magnetar Read: Build the Product Around the LLM
[47:25] — Flip It: Human on Idea Gen, AI on Risk
[50:42] — Advice for the AI-Native Analyst
[54:34] — Using AI to Deepen an Experience, Not Skip It
Want to actually build these workflows yourself?
The AI Accelerator is Fundamental Edge's 6-month cohort for investors who want repeatable AI workflows. Learn More below:
https://www.fundamentedge.com/ai-accelerator
Watch the full podcast series on our site: https://www.fundamentedge.com/invest-with-ai
Follow Invest with AI on:
Spotify: https://open.spotify.com/show/033xcEEovVViS7hIYwNuGZ
Apple Podcasts: https://podcasts.apple.com/us/podcast/invest-with-ai/id1896918892
I think the ROI, I think the value add of a junior analyst has the potential to be much higher than it was in the past.
SPEAKER_02And so as I read that article, I was like, they kind of took LLMs and said, let's build a product around the strength of this thing versus let's take this thing and put it into an existing product structure.
SPEAKER_00It's a great business. I think it's the greatest job of the world. With the introduction of AI, I think it's insanely exciting now. But especially for a firm like ours, which is really lean.
SPEAKER_03Welcome to the next episode of Invest with AI. We're really excited to have Raj Shaw, who's a co-founder of Stoic Point, on the pod with us today. Raj is a brother on the journey. I first met him in analyst training at Green Hill in 2007. We both walked our own hedge fund journeys together. And now we're sort of walking the uh the AI journey together as well. Me more on the sidelines, Raj in action in the arena. So uh really excited to have you, Raj, today. Just sort of chop it up about uh AI and investing and uh and what you've seen as useful in your investment process. Do you want to sort of elucidate your background or tell us a little bit more about sort of uh how you're coming at this question?
SPEAKER_00Yeah, absolutely. Well, look, I think you hit the earliest part of it, Brett. I think you and I were the only two in our analyst class of nine or ten to leave early to go uh to go join hedge funds. Um and I think that was a generally a a good uh a good decision. But uh I spent the first uh after that the following five years at a firm called Highline Capital in New York, and then I moved out to join Light Street as the third partner in uh Palo Alto. I've been out here in Bay Area ever since, and my partner Cullen and I started Stoic Point in 2018. Um kind of been bootstrapping the business all together, completely fundamental, research-driven, concentrated portfolio. You know, our median hold periods are in years, probably two plus years, almost three years on a median hold period basis. And we've got you know 10 to 15 names in the portfolio, of which I'd say probably at any given point, eight are kind of eight or nine matter. Eight, eight to nine names are probably 80% of our exposure. Um, so we tend to do a lot of deep work up front. We tend to um invest in these businesses for a long time. We tend to engage with these companies, um, management teams and boards alike to sort of help them navigate their paths as as public companies, uh, generally focused on small and mid-cap, and we'll do anything other than really energy and and biotech. It's kind of two areas where we'll say, look, that's not an area we feel like we could ever be the smartest or near smartest in the room. So let's avoid those altogether. Um we are, for all intents and purposes, kind of a two-man operation. I mean, we've had um outsourced analysts, we've had juniors, we have some interns this summer with us too, but but we like to keep it lean, not only on the portfolio side, but also operationally. Um, the hedge fund business model for as uh as much ink has been spilled about how hard it is to operate a hedge fund to start one, to keep it going. Um if you can run it lean enough, if you can build for longevity, if your the life of your fund can match that of your investment time horizon, it can be a great business. And so I've had a ton of fun doing it, not to say it's without its, you know, it doesn't have its stressful points in time. Uh, but um, but it's it's a great business. I think it's the greatest job in the world. And um, with the introduction of AI, I think it's it's insanely exciting now, but especially for a firm like ours, which is really lean, where you have portfolio co-portfolio managers who are also at their core analysts. Um, and so we get to really just pour a lot of fuel on our research process and try to get a lot more done a lot more efficiently. Um, so it's been an exciting time. I'm excited to chat about the AI stuff because it's what we're spending a lot of time on.
SPEAKER_03Yeah, yeah. No, thanks, thanks for that. And I think we've sort of both we've both seen inside large funds of the resource stack that is wrapped around an analyst, and it's it's phenomenal, like any resource at your fingers. And when you leave a firm like that, you have to recreate that resource stack. It requires a lot more hustle and creativity. How have you tried to solve that problem at Stoic Point initially? And just walk me through your sort of discovery journey with AI and helping solve that resource stack. Like, how do you bring that institutional level of analysis to a firm that doesn't have the resources of a $5 billion hedge fund?
SPEAKER_00Totally. I look, I think I'll I'll make the first part of that question, I'll the answer rather quick, because I think it really accelerated in the last you know year or two with AI. But look, I think there's a big benefit in this industry to put it bluntly not being a dick. Uh and so when you go out on your own, um, you uh will tend to find if you have forged good relationships at much larger funds and working with the right people, whether that's sell side or service providers, people will tend to invest in you and give you access to resources for you know a pretty decent amount of time as you sort of get ramped and people are either emotionally or in sometimes uh in some cases financially invested in your success. And so we benefited a lot from that uh early on. But it's um it's been um it's been an extremely exciting uh point in time since kind of you know, chat, it'd be probably three months after ChatGPT first at the scene in late 2022 to apply a lot of that to our process here and get a ton of leverage on what we do. For us, the reason we get up and do this every day is because the most fun we have is is finding new ideas. It's that research part of it, it's that early part of the process of I think I've got something here, I think this is an interesting idea, I'm still doing work on it. I've I got a hundred questions I need to get answered, and I've got to work out a plan to get them answered in the fastest way possible. I also, by the way, want to meet the company, I want to ask them these questions face to face. I want to understand what the street thinks about this business, about its valuation, about its outlook. And so there's just a whole bunch of things to go and do. And when you work at a large, um, at a large and well-resourced firm, Brett, like you were alluding to, there's other resources you have. You might have a director of research, you may have a junior analyst that can go and chase down, you can sort of divide and conquer. Um, but we don't have that. And so for us, time is really the most expensive currency. It's there's return on time. And AI has really been an unlock for that. Um, when I break out and think about AI in our investment process, I think it's kind of in three primary buckets. One is that screening and idea generation part, the part that's really exciting is finding those interesting, out of consensus ideas. For us, it's those messy situations we like to say that we run towards the fire. AI is obviously very interesting in that application. We'll talk about that. I think the second one is that research of an idea part of it. Um, and that's where I think a lot of people are spending time talking about how do we really use AI in this process. I break it apart into there's like the sort of deeper research due diligence, what people call due diligence level part of that. But then there's the basic blocking and tackling. I gotta make a model of this business. I want to sensitize a couple different cases. I want to create some output that maybe other members of my team need to consume or my investors need to consume to make the case for why this is interesting. That's kind of part two. And then I think the third part is is the is the idea or rather the position maintenance, position monitoring part of it. Once the position makes its way into the portfolio, there's the ongoing monitoring of relevant data points. And I think AI touches um all of those things. I think part one, that screening and idea generation part, is the part we've we started out the most excited about. We've continued to be the most excited about, but I'd say it's probably the piece that's in a lot of places sometimes the most disappointing. Uh, what I mean by that is there's there's a lot of places where we're the idea here generally is there's you know thousands of potential investable ideas. There's new companies coming public every day. The IPO window is open now again, so there's even more companies uh in this universe being added every day. And so how do I triage that and how do I find where I'm supposed to be spending my time? One way I've always done it, and I think a lot of firms do this as well for good reason, is you run screens. There are very simple screens and there are very complicated screens. Simple screens would be hey, I tend to find broken IPOs really interesting. So let me just go and find every company that's IPO'd in the last two years, it's down 20% or more, or 30% or more, and I'll go dig into all those. That's a simple screen, that's a quantitative screen. You could do that on Bloomberg, you probably do that on Yahoo Finance if you needed to. Then there's the more interesting, sort of nuanced screens that I think AI holds a lot of promise with. Think about turnaround situations. Turnaround situations can be very hairy and very messy. I want a company that has a new CEO that's coming the door. Well, new CEOs usually need a year plus to get their, to get their sort of hands-on a business. They may come in and cut guidance and the stock will fall. They have to change out the management team. There's a lot of stuff that happens. So actually, I want a company that's in a turnaround where the new CEOs come in and he or she's already been there for a year. I also want the stock to have been down. I want stock to have been down 20% or more. But I want there to already be evidence of the turnaround working. I want them to have, you know, beaten raised for the last quarter or two, maybe you know, three, four quarters into the CEO's tenure. And I want street estimates to start going up in some signs of life. That's where those types of things, when you can start to get a little bit turn these cranks, that's where AI offers a lot of promise. And we've seen a lot of interesting tools offer those types of screening capabilities. And excites us a lot. What's been somewhat, I not even, I wouldn't say disappointing. Disappointing is probably a strong word, but confusing is LLMs are a black box. And then when we've run some of these screens and we've run them repeatedly, or I'll run it, my partner will run it, we'll hit click at the same time, same criteria, exact same prompt. We'll get like two different sets of companies. He'll get a dozen companies that arguably fit the criteria. I'll get another dozen companies that arguably fit the criteria. And we're looking at each other's results and we're like, same prompt. Why didn't why did we get different things? And you go back and you try to figure out why. And again, LMs are black box. So it's it's it's it's hard to figure out. Um, but nonetheless, it's still really exciting. Um, I think the other thing that's really important that we've been learning on again on this idea generation and screening part is the AI is only as good as the data that that it's that it's using. And so for us, we tend to gravitate towards, I mean, special situations are like you know, one area we love. So transformative MA, for example. Transformative MA is messy. Generally, street estimates, financials don't end up getting updated well after a deal is closed. And so there's this really interesting period of time where if you've done the math on what shares are being issued, what's the pro forma balance sheet, who's running the business, what does the growth look like. There's a great opportunity to get in there early at transformative MA situations. But LLMs are not necessarily going to be great at that yet if they're relying on SEC filings and that type of data in identifying those situations for you. What it does hold the promise of is I should be able to put a slide presentation about a merger into the LLM and have it produce the pro forma financials and flag for me, hey, by the way, this company is now, after a result of this deal, the cheapest among its 10 peers. That's pretty interesting. Synergies look really reasonable compared to the OpEx at you know A B and C peers, those types of things. And so we're really excited about it. I could go on and on about the screening and idea generation front. Bottom line is I think it's still the most exciting part of this whole AI in investment management or hedge funds world, but it's also the part that I think needs the most the most work.
SPEAKER_02Can I ask you about the the screening? So the two-prompt question. Such a good like uh like A-B test right there. Can you say a little bit more about how that is set up? Is it you know a really good chat GPT prompt or skill that you've reused, or is it you've mentioned third-party software? Like, how are you actually running that prompt?
SPEAKER_00So we've um we've done it a couple ways. And it's a great question. Originally, we started doing we started using some service providers that um that purport to have these types of tools. And and in a lot of ways, they're they're really good. I shouldn't say purport to have them. They have these types of tools. Portrait is a great one, and I I know David's been on with Brett before too. Um, Portrait's a great tool that can take a lot of your natural language queries and turn them into both quantitative and qualitative screening, and then it'll go and scour and find those. Um, and that's great. But we found, you know, I I should say I think LLMs almost always, they're they're always going to try to find the shortest path of the shortcut to getting you the answer that you've asked for. And so I think part of the problem in why you get different results is that when you put this criteria in, it's sort of scouring only part of the investable universe. If we do a similar screen on Bloomberg, for example, using EQS, EQS is like the OG screener. It's actually very, very good. It's in it's almost entirely quantitative. But if you run that screen over and over, you will always get the same results back. Like I have a very high degree of confidence. When you run an EQS screen, it's it's it's it's finding everything and everything is accurate as much as you've put the data in. So what we've started to do because of some of these sort of LLM native screening tools have had these inconsistencies, is we'll sort of start with Bloomberg. We'll start with the quantitative to know that we've scoured the entire Investible Universe of these ideas. We'll spit that data out. Bloomberg's kind of been the sleeper, I feel like, in the AI wars for hedge funds right now. Because obviously they sit on all this data, but their map their models have generally sucked, but they're they're coming from behind here. Bloomberg's now, because their language, because it can dump into Excel, you can take those screens out of Bloomberg and then throw them right into somebody else's product and say, look, this is the universe I want you to start with. Don't worry, I don't mean to pick on portrait, let's say Alpha Sense, don't worry Alpha Sense, don't worry portrait about burning a bunch of tokens running this screen for me. I've already picked from 1,500 names, I've picked 500 for you. Now go do this qualitative screen for me. And so I think tying a lot of these things together helps um helps on that idea generation front.
SPEAKER_02And Brett, we talked a little about this with Ying on the podcast. It's like the separation of the deterministic query from the non-deterministic query. And this is Raj is giving a perfect example of you've kind of hacked a way into it, right? But eventually you would imagine that those would merge.
SPEAKER_00Yes. And we've we've we've hired a couple enterprise and college students this summer, and this is their core focus is hey, we know there are some strengths here, there's some strengths here. Like you said, deterministic, non-deterministic. Let's actually just build an internal solution where we can weave all of these things together. And they've only been at it for like a week, and it's showing a lot of promise. It's also showing some obstacles because it's funny, some of these resources that we're using that we're paying for really would rather prefer to be walled gardens where you're doing all of your research. And they make it really hard to sort of leverage the best of what they offer and tie it in with the best of you know what somebody else offers.
SPEAKER_03Yeah, how I it's a if I got a sort of interesting conversation there, too, of like how how are you thinking about the the UI here, right? It's like in the past to try a bunch of these chatbots, you had to have six logins. And what's the new dashboard as we sort of transitioning from chatbots to agents? How are you thinking about the UI and how all these things plug in together? It still feels to me very disjointed. What have you done to sort of turn this into a new seamless operating system?
SPEAKER_00It's it's it's terribly disjointed. It's a huge problem. In some cases, you're like wrestling with PDFs, in some cases, you're wrestling with spreadsheets, in some cases, it's like natively text on a website where like your best option is to like export it. And this is, I mean, we haven't the answer, the short answer is we haven't built it yet. It's it's it's like goal number one for the next two weeks for you know our enterprising college students is to figure out the best way to put this together. Part of the interview process we had given them was actually to go filter through a bunch of names and like do some preliminary screening for us. And one of them came back with a pretty interesting output that they used Claude Code to create, which was an interactive website. And it had pulled in using MCPs, using, in some cases, agents to like navigate with a mouse, some third-party website, and pull that data in. I think that's ultimately probably the way we're gonna end up consuming this, is but I'd like to be, I don't know, I'm a you know, now an old hedge fund portfolio manager. I'd prefer using my keyboard over my mouse. So I'd rather not have to click a hundred times to go get to the data I want. But I think the reality is it's probably gonna be something like a you know, a very simple website where every day it's filtering in. I we originally thought of emails too, like just email me when you find things. We have this concept of we internally we call it the meta screen, which is like we have a dozen screens, we probably have 50 screens. If there's a company that shows up on like seven of those screens, like slap me in the face with it. It's it's on the meta screen. Like we should absolutely be looking at that scenario. And still, though, if it emails it to me, like I'm probably still gonna get it, like I won't, I won't get to it till later. You know, like it's just gonna get lost in the box. So I have to establish this workflow of that web page has to be one of the first five things I do every morning. Just go check that site, anything that's popped up that looks really interesting. So that's what we're theoretically playing around with right now, but you know, this is this is evolving. And I know maybe it'll be an app.
SPEAKER_02On that, on that point, Raj, the the Claude Code project that that the intern built, you you mentioned it just to call it out for others, the computer use has gotten really, really good. Like, especially on codex, where it full on just takes control of your computer and clicks through blue. Like, and um it's funny because some some people believe it's like why even bother with APIs and all that because computer use is getting so good that we're just gonna like give AI this human thing, but they're good enough at it. So it's interesting to hear how you're no, it's totally true.
SPEAKER_00I think it's a matter of speed. Like if you have a direct API connection, obviously you can just move faster, but you're absolutely right. Like it's it's contextualizing these pages, it's clearly remembering how they work, so that the second or third go around, it's it knows exactly where to go to pull the information it needs to. Um it's it's scary, but it's I mean it may be the way we sort of to Brett's the way Brett, you kind of have to hack it together. Can it can this go for now? I don't know.
SPEAKER_03Like, so could codex computers today go into my Bloomberg, run BQS, download the Excel, go to Alpha Sense, pull in expert network calls. Like, can it does it have that capability?
SPEAKER_00So it can do almost all of that except for the the Bloomberg part, and this is like probably somewhat by design, or it's just kind of been this way and it's worked out fortunately. The Bloomberg authentication process is obviously like notoriously annoying because you gotta do it's 2FA. So if you're logged into Bloomberg, yeah, absolutely, it can do all of that. It can grab stuff from Excel. Claude, by the way, I don't I actually don't know about Codex2, but I would guess it would be Claude is remarkable at writing Bloomberg like BQL language. It I don't know if it's ingested all of the documentation because BQL is like still fairly new, but you ask it to write an entire script and it'll do it in BQL in a matter of minutes. It's pretty amazing. And so you can do that. You can do that with co, you know, with cowork, or you can do it natively with Claude in Excel and it'll drop the data in for you really quickly. It'll export it. You can then manipulate, you can upload those in the Alpha Sense. Um and that's part of what we're doing, is going back to this like specific providers and how we're sort of piecing this thing together. One of the things we wanted to do to kind of just jump ahead here, because I talked about sort of point three of AI being monitoring, portfolio monitoring, is we we don't have nearly as much turnover in our portfolio, I think, as the average fund does, but we do have some turnover in our portfolio. I want to be able to, on an automated basis, if a new name comes into the portfolio, I want the next five to seven comps of that name. Automatically, I want to start monitoring news flow for those names. I want to start seeing if there's anything on expert networks coming off. So what we can actually do, and we're working on this in real time, is um you put that, you have that our our portfolio management system or our order management system will automatically generate a file for what our portfolio looks like. That can then go into Claude. Claude then threw an MCP into Portrait, which it has direct access to, will then ask Portrait, hey Portrait, you know, you have a much better idea than even Bloomberg who the next five comps or peers to this company are. Give me those next five peer companies. Claude can take that, throw that into a watch list for Alpha Sense, which then has expert network access, has access to SEC files, and has access to news flow, and create a monitor automatically for me. And so again, it's like piecing all these things together, but it gives you at the end of the day, for a two-person shop, an institutional level ability to monitor a lot of positions and know what's going on, uh, not only in your names directly, but what's going on maybe in their industries or with their competitors. We had this great thing. We own a company called Lux Experience. It's like now the dominant e-commerce provider for high-end luxury. It was the result of a transformative acquisition, which is what got us involved in the name. And one of the things that was behind our thesis was look, this is kind of gonna be the only place that these luxury brands have left to go if they want to do e-commerce outside of their own websites. And we'd done this process, and like three weeks after we put this process, this like, and it was a light version of what I just described in. I got an alert from Alpha Sense that Brunello Cuccinelli, the luxury house, had specifically called out Lux Experience on one of its commentaries as being one of the biggest growth avenues for its business. I'm like, that's huge. I would never have put a you know random Italian shoe, you know, shoe company on my monitor. Monitoring, but Alpha Census AI sort of knew to do that, and Portrait had sort of told it look, this is one of the peer companies you should monitor if you care at all about luxury. So it's it's it's pretty amazing, like what you can do. I'm sure if I worked at a much more well-resourced shop that had analysts, that maybe there was a guy who was covering luxury, and then I'm the guy covering internet, like he maybe he would have told me, but you know, that's obviously not the way we work.
SPEAKER_03Yeah, fascinating. Um tell me more about the um the interns, a sort of a uh narrative violation to the current narrative on the street that AI is killing juniors. Like, how have you thought about the ROI? Obviously, sort of a uh uh discrete cloud code projects, you know, a nice thing to have interns work on. But how do you think the ROI of junior talent shifts? And how have you thought through sort of the utility of of uh of you know uh younger, more uh invest you know, AI native uh perhaps investment talent?
SPEAKER_00I think the ROI, I think the value add of a junior analyst has the potential to be much higher than it was in the past. And I agree with you, I think it's a narrative violation. Like it's so easy to say they're gonna get run over by AI. Look, I think I'm much more likely to get run over by AI. I'm a 41-year-old portfolio manager who's like rooted and learned the ways of the past in terms of how to invest and how to do research. And these guys are coming on the forefront. They know the best tools to use when a problem is placed in front of them, and they have a real potential to be one of the most valuable people on an investment team because the world is changing. And so there is now every day more places to go to get information. It doesn't just reside in Street Account and Factset and Bloomberg and CapIQ, like it's everywhere, and these guys know where to go get it. So anybody who is creative, who is driven, can enter an investment uh firm with a whole pipeline of ideas of, hey, here are differentiated ways that I can accelerate our workflow. Here's I can get more coverage and more efficiency out of us. And so I'm excited, I'm super excited about it. You know, when we started the firm, and even as we've sort of checked in as we've grown AUM at like where we want to invest resources, we originally had in you know, an analyst at a higher threshold of where we were gonna be. And now I'm like, man, and we were originally gonna hire someone in the back office first. But now seeing what's possible with somebody who's sort of just an AI native person, there's a lot of talk about AI native companies, like an AI native person, I like that value of that person is extremely high potentially. And so that's where the next hire for us may very well be is we're gonna hire somebody who can come in and automatically start knocking things out that we didn't think were possible. Um and it's different. I would sort of add it to you, Brett. Like I think on fundamental edge, you know, you started out, hey, I'm gonna train analysts in the ways of the tiger cubs and the ways of the platforms, you know, how it's been done for the last 10, 15 years, the way you did it. And then I think you very early on saw this happen. You're like, well, shit, this is totally gonna change now. What the analyst needs to do, not only to survive, but to thrive, is completely changed.
SPEAKER_03And um, I think it's exciting. I'd sort of give you one I'm giving to all PMs is uh you've taken so much time to articulate your screening, your research, and your monitoring process, go into your systems and be like, please turn this into an intern training guide and just have it spit out a 30, 40 page. Because like most PMs haven't done that, they're now doing that almost for the first time, deeply articulating all elements of their process because that's the key thing that makes these agents go, right? That also creates an exoskeleton for the junior person no longer has to learn through osmosis, they can actually see that articulation, they can step into the exoskeleton and actually spend their time in a much more effective way with human action cues and all sorts of these other ways.
SPEAKER_00And they can query it, which is amazing too. They can ask, yeah. I mean, the analog to that is like just our investment values, our investment process. Stoic Point has three core investment values. It's how we sort of measure everything. We have our own investment scorecard, we grade all of our investments that way. And in the past, I'd have to sit with somebody and sort of walk through that. Instead, we upload it and we write, sadly, like very long investment letters. We don't, we only now do two a year because they're so long, but we've written hundreds of pages of our thought process, what we like about ideas, what we don't like about ideas. Some of those ideas in there are probably not good ideas, but nonetheless, we've sort of detailed them out. We can now, you could throw them as simply as you know, somewhere in a notebook LM. And a would-be analyst or an analyst that we're going to hire can go in there, listen to a podcast about how we do it, read, you know, get summaries of different ideas, and is there you know, specific ways or mental models that we use, and they can have a conversation with an LLM about our investment process. How cool is that? Oh, yeah.
SPEAKER_03Or do uh uh I've seen funds do like the the Raj agent where the intern has to pitch up to an agent first before I guess they you know find find the logical flaws in this thesis, yeah. And by the time they sort of sit down and pitch you the idea, some of those dumb mistakes have been ironed out of the of the process. It's pretty good. If you haven't done that, like I love that. And what we have to do is like actually record a handful of those. So it's like, you know, obviously the compliance record a few where you're just ripping apart, ripping apart an analyst pitch, and you know, the agents can infer that pretty effectively and turn that in turn that into a sparring part spar sparring partner. Um I like that a lot. Grounded in your creditors, etc.
SPEAKER_00Right, right.
SPEAKER_03Uh talk to me about um um Excel, which has been sort of a challenging one. I mean, to sort of today, even still, you know, Microsoft Excel is the workbench of the fundamental investor. Sort of like what we do is so heterogeneous as investors, there's been no front office software. So like Excel's been it. Um the translation layer from research to decision. Um what have you done with AI and Excel and sort of where do you see that? Where do you see that going?
SPEAKER_00I I it's probably of everything we've talked about so far, it's probably been the most um, it's probably been the biggest positive surprise of everything we've used in terms of AI applied to hedge fund investing and managing business and all of that. Um, we started using out, like I'm sure many people did, a lot of the startups and a lot of the guys who were early to doing AI integrations into Excel to help sort of turbocharge model building analysis, things like that. And then very quickly just sort of gravitated towards Claude, Codex, those integrations with Excel. And they're they're fantastic. They're amazing. Um, and they've sped up on the f on the good side, they've sped up our ability to just cut data different ways to create output. Um, a very simple prompt everybody should use. It actually speaks to what you were just saying, Brett, about the pitches. Is you can just ask an agent, you can ask Claude when you have one of your models open to test this model. Like, where is the structural flaw in the logic of my model? Yeah. And it'll, it's actually quite insightful. Oh, wow, the way you're you know doing free cash flow conversion makes no sense at all. Or you're building a bunch of cash on the balance sheet. Is it gonna be used to buy back stock? What are you doing here? Sort of a lot of times it's funny going back to like when I used to look at sell side models, I'm like, oh, that's dumb to sell side model it this way. Well, they do it for a reason that way, but oftentimes like it points out those dumb things for your own models. And it's a great sort of check privately, you know, it doesn't embarrass you in front of anyone in terms of where a flaw in your logic might be. You could then take it a next step and you can say, hey, here's the last three transcripts, here's some appearances at conferences, here's what peers are saying. Is there anything about what the company's articulating about its near-term to medium-term to long term that's inconsistent with the way I've modeled it? And the the combination of a smart LLM with the ability to navigate and contextualize Excel is amazing. And so you can use it that way on financial models. I've loved doing that because the modeling is look, as I've gotten older and I've become a portfolio manager and a founder of a fund, I spend less and less of my time in Excel doing model building. And so it's been helpful to help me just remember look, here are the things that are important. But there's all sorts of other ways in which I've been using it. We've, I mean, very simply, we had a uh one of our largest holdings, we were trying to help them understand the difference between talking about growing at a lower rate versus talking about growing at a higher rate and the valuation gap between those lower growth companies and the higher growth ones. And very simply, we were able to create a scatter. We said, look, dump a hundred companies in there, dump 150 SaaS names in there and spread apart all of them by market cap, spread apart all of them by um growth rate, cut out the low gross margin ones, a ton of different analyses. You just keep firing into Claude in Excel. It's and it's just repeatedly doing a bunch of analysis in real time for you. Two years ago, I started to become a more power user of Power BI. We were originally doing Tableau and things like that to try to do data visualizations. And I found myself like drawn right back into Excel. Excel with Claude, you know, next to you is kind of better than doing a lot of these data visualization things because it's newer. I hope maybe they can get you know a good LLM into Tableau or Power BI, and then maybe then you're like really off to the races. But it's been super instrumental in that. And then the last thing is like just tell me a story. Tell me a story about this set of data that I've that I have here. Um I'm it's funny, I'm it's totally unrelated, but I just I just bought a new home and I'm in the process of selling my old home. And I had a meeting with my real estate agent who's selling my uh old home, and she had a bunch of you know insights and like, you know, that's like more art than science. I'm like, hey, here's where I think you should price your home, and here's a bunch of comps, and here's a bunch of listings. Immediately threw all that stuff into Excel. And I was like, hey, what do you, Claude, what do you infer about home prices relative to other characteristics about these homes in the town in which I live here in Northern California, which is Northern California is like an insane place for home buying and selling right now. And it very clearly drew a correlation between the, in in this specific town, between the age of the home and where homes were being listed at. And it was basically inferring that look, people tend to in this town, people tend to bias towards homes that are newer and they don't have to be fixed up as much. I'm like, that's an amazing insight for quad in Excel to just do. Now apply that to all sorts of other sets of data, not just models, not just the boring stuff, credit card data, other kinds of alt data. Inferring patterns is is, I mean, like such a great use of this.
SPEAKER_02Hey Raj, on on that point, you tying that to the intern and the junior question, you're deliberately not trying to hire young, you know, more junior folks given it's trying to stay lean. What happens though if larger firms do what you just described? And you're like, well, we don't need the juniors to do the models because everything you just described.
SPEAKER_00Yeah.
SPEAKER_02What do you what happened like do you do you believe that there's a pro a challenge for like the next layer of talent, or or is that also kind of a narrative that we're falsely buying into?
SPEAKER_00I think it changes the progression of a junior analyst and not necessarily in a bad way. So I think the junior analyst now has to be sort of AI native and AI fluent as soon as they come on the job. Whereas maybe before you just had to kind of be Excel fluent. And you still don't have to know a lot about what makes a good business or a bad business or how to read a CEO and how to think about consensus. You learn all of that stuff on the job. I still think you're gonna need to do that. Um and so I think the junior analyst will still grow into those types of things. I still think it's gonna necessitate those types of conversations. Like, hey, I created this output, I've done this analysis, I've done the screening work, but let's also have the conversation, PM or sector head, about why you think this idea is interesting, what makes it a good business, what time horizon we're thinking about, what are red flags we should look for. The analyst is still gonna need to know all those things. And I think that goes back to this point that a few people have made, but I don't think it can be said enough that AI will almost always lack that judgment factor. And that's the part that junior analysts will grow to learn. You learn through experience, you learn through uh exposure on the job. I I kind of hope for my own sake, because you know, I hope my competitors just lean too hard in the AI stuff, that the judgment stuff sort of becomes lower priority and it'll present better opportunities for me. But I think the reality of the way this industry is going to evolve is people are still going to lean on the judgment side. We've just now sort of embarked on a whole new set of tools to make us more efficient. That's the way I kind of think about it.
SPEAKER_03Yeah, that uh that makes sense. So the the the hearing interesting rumblings about LLMs sort of developing judgment in various tech various tests and competitions, too, which starts to get sort of scary. Yeah. Sort of thing. Think about like if all of a sudden, you know, LLMs could actually get it give you a 55% batting average on earnings prints, uh that's sort of a totally different conversation.
SPEAKER_00Like scary, not to say that's my base case, but I'm hearing more people talk about that as a But it's funny, like right, Brett, if you play that out, it if if enough LLMs do that, then it sort of already defeats the person. If enough people are using an LLM that's giving you sort of that same answer, then you won't get that earnings move, right, after the print because everyone's sort of I feel like it's kind of akin to like alt data started creeping out, right? Whenever that was 10 years ago in the industry, and the first early adopters of it were actually generating some alpha. And then after that, you had these companies like beating comps and the stocks were going straight down. You're like, God, why is that happening? You're like, oh, well, it was happening because everybody's looking at the same set of data. It's probably a similar analog to like what could happen if people become too LLM dependent. Yeah. I think the the most amazing, sometimes humbling thing about this job is the fact that you can be right and still lose money or be wrong and make money. And I don't think that's gonna change, you know, with this.
SPEAKER_03Yeah, yeah, yeah, exactly. How do you think like um, you know, uh we do uh we do a we're doing AI trainings for enterprises, and one of the questions we get is like, how do you measure the ROI on your trainings? I'm like, well, ROI has to be measured in PL, right? Like that's ultimate, that's the law of the jungle. Uh in the hedge fund world is make money. How do you think about measuring, you know, it sounds like you spend many, many hours on this over a number of years now. How do you think about measuring ROI for all of this motion?
SPEAKER_00Yeah, look, on a purely quantitative basis, and it's gonna be hard to disaggregate, right? Like the PL that was attributable to AI initiatives and AI work versus what was sort of just done otherwise. That is, I think, clearly a quantitative way to do it. I think it's actually if you're a portfolio manager or running a fund and you have multiple demands on your time for different parts, like I've got the part of the business side of things that I have to spend time on. I've got the fundraising side of things that I've got to spend time on, and then obviously I've got the research and portfolio management time. For me, the ROI is going to become pretty evident. Either though the ROI or lack of ROI is going to become pretty evident if I can find myself spending more and more time or higher output out of the time that I'm spending on research. And so if I, and we track this, we have an RMS research management system, ideas go in there. We're actually trying to automate that too. We're like not nearly as close as I want to be, because I want to connect everything we just talked about with our RMS, take ideas from Idea Gen and put them directly into the RMS and track their progress. But if at the end of a year or six month period, I can say, wow, compared to last year, I processed 25% more ideas than I did, you know, the quote the year before, a year ago, or a quarter before, that to me is going to be a positive ROI. I may not have actioned all of those ideas. Hopefully I didn't, because I we don't tend to try to turn positions over that much when on these for a while, but I would have processed a bunch of ideas, then that will feel like a good use of the that investment of both time and capital for me. Um, because every new idea has to compete with an existing idea in the portfolio to get in. So if I can just keep going through things and figuring out, hey, this does meet or doesn't meet and just fast fail them, I'll log all that in my RMS and I'll have a data set that I can look at to tell me, hey, how many things did you did you look at in your investigation?
SPEAKER_03Yeah, that's a great, great data point to to measure. Um I want to get Kay's perspective on this too. But as you thought about as a former allocator, uh as you've thought about raising capital and telling the stoic point story to to allocators, we've sent been sort of this like ascendancy of the multis, the sort of uh, you know, uh it's been a tougher environment for the single managers to raise money. You start to think about this new resource that helps smaller funds do more. How have you thought about those conversations? And do you think this will sort of be a broader trend of sort of the sort of ascendancy again of smaller investment teams really hunting for alpha outside of you know the 10 biggest stocks in the world?
SPEAKER_00Yeah, I do. I absolutely do. I think um on the we absolutely talk to investors about our ability to get more done, to get to sort of be resourced like a fund that's three, four, ten times our size now, thanks to these tools at our disposal. And that's that's part of it, that's great. Um, but it can cause you to like make a bunch of unforced errors if you're just like doing a bunch of stuff for no reason. Um, for us, the way we think about it and talk to investors about it more from the macro standpoint about this, like as you said, a lot of flows have gone to the platforms, and do we see that flow, those flows come back to single managers? That's going to be borne out in performance. And I do feel like we are at the dawn of probably pretty exciting period of time for managers who do something similar to what we do, which is you tend to keep a cooler head among more volatility, you tend to not measure things and measure things in shorter time periods and you know, data points, uh, and you tend to have a longer uh time horizon and duration. And if you have those things, like the quality AI, the at you know, the appearance or emergence of AI in investment management hasn't changed the underlying businesses that we're all investing in and looking at. All it's changing is the amount of analysis and compute directed at those businesses, potentially increase the number of data points that are being looked at. But if you're ultimately have conviction in an idea, the business you're investing in is working, or the short you're putting on is destroying value, a hundred or a million more data points isn't going to change that, you know, ultimately, but it could lead to a ton of volatility in the stock. And so I think for people with cooler heads with the ability to add on those drawdowns who have done pre-mortems to underwrite, hey, here's what I would do in those scenarios, or can trade effectively around those, you can generate a lot of excess return that way. And so that's generally how we've been thinking about it. Is I mean, short way to say it is we think it's gonna just increase volatility amidst all the other things that are happening. I think 24-hour trading and a lot more options activity in the presence of retail, volatility is just gonna continue to go up, I think. And I think that will benefit us overall. And I will hand it to the just general, and this is a vast generalization, but allocators overall, I feel like have become more comfortable with volatility because they're saying, look, I care more about getting from point A to point B, and the in-between is maybe less of a concern to me. And so if that's the type of allocator that you are, I think you should be looking at a lot of single manager hedge funds who are thinking uh, you know, about things this way.
SPEAKER_02Yeah, you know, I I've been removed a bit from the direct allocator space. So if I if I go back to put my hat on, I think part of the at least the large allocators are probably still struggling to understand kind of all the secondary risks associated with AI just in general. And so that might be uh I would say that might be a near-term challenge on like, hey, we like we don't have a you know uh a director of ops because we use you know it's claude and our our head of IR is is Chat GPT. But I think that like anything at the end of the day, as Raj Raj said, performance over a cycle performance in that, you know, if if it changes the cost structure of things, you know, that will then bleed into performance on a very clear kind of one-to-one basis. Like like people have been saying the cost structure, the the performance, the compensation structure of a hedge fund is gonna change for 20 years, and it hasn't really. Uh so it'll be so there'll be a bunch uh of different forces, but it does, it reminds me a bit of you know, the early, I'm gonna I'm gonna date myself, but like the 2003 hedge fund, like you know, there's the two guys in the Bloomberg, right? Where it's you know, that died, you know, that that took a little bit of a break, but um it probably has the the tailwinds to come to come back structurally.
SPEAKER_00Yes, yes. Look, you've made it so much for so many decades, has made it harder and harder to operate a hedge fund. The regulatory costs, all of that um have driven up, but now you may have a lot of forces working the other way. And look, I'm out here in Silicon Valley, and a lot of people are talking about, especially as there's waves of layoffs going on. The silver lining is all of this, is we expect to see even more entrepreneurship. People are going to start businesses quicker. You know, the next billion dollar company is going to be a one or two person shop. I think there's a parallel there in the hedge fund world too, where we could return, as you said, Kate, to the two two guys in a Bloomberg. Maybe it's not even a Bloomberg anymore. Bloomberg's really still fucking expensive.
SPEAKER_03Two guys in a codex, uh, two guys in codex uh accounts.
SPEAKER_00That's right.
SPEAKER_02You know, can I run something by by both of you? Because it just hit the tape like you know a week ago. So, and I have no inside information on this, but Magnetar announced that they were relying like more heavily on uh AI to do the early part of their research. But what I've always found interesting as kind of more of an outsider than both of you is that I think you have to change the business model and the product as well to adapt. To LLMs. And so I read that article. And again, it's just off of one article. So it was a lot of speculation on my part, but with intrigue. Because a few things that they picked up on that I picked up in the article. One is that they were they really emphasized that they're going to use LLMs for the early screening process and human judgment at the portfolio management side, like later in the process. So like picking. So that was interesting because it's like that separation that we've said of you know data gathering and synthesis and judgment. But there were two other things. I would love to hear your thoughts. One is that they made it concentrated long only. So they basically took very natural attributes from a business model and kind of like layered uh AI into it. And like structurally, there's a good trend. Like if you're a good stock picker over a longer horizon, you've got a tail, you've got the market risk premium at your back as well. And so as I read that article, I was like, they they kind of took LLMs and said, let's build a product around the strength of this thing versus let's take this thing and put it into an existing product structure. I'd be curious if you agree or if I'm reading too much into one Bloomberg article.
SPEAKER_00I think Brett, you can you can go first as some thoughts.
SPEAKER_03I think it's sort of just like harder to measure skill in a long only sleep, which is sort of sort of hard. So I mean I I would probably like start in a different way. Like I sort of started with like measurement of like batting average on earnings prints, which is just sort of more sort of easier to track sandbox. Um, and you don't have three bad earnings seasons by accident, right? So you can sort of know within nine months whether the thing has skill or not. Um at least it's sort of the multi-manager DOR hat. Like we'll give you one bad earnings season, two, two, you're on watch list, three, like you're out. Um, so I think you know, that's how I've sort of thought about like a good sandbox for judgment, where long only you're sort of so driven by like you could obviously sort of factor neutralize the idiot. Um, but I I feel like it's just gonna be a very hard measurement, very hard measurement uh problem to sort of judge whether there's actual like excess performance in that uh I think generally like you're seeing a clear trend of convergence. You're seeing the quants sort of use these tools to become more fundamental and the fundamentalists using using this tool to become more quant. So like I think the the concept of an investor, right? There used to be even like 10 years ago, I'm a tiger guy or a multi-guy or a quant or I think like the concept of what an investor is is going to look like more similar than different. Um yeah, you know, very very soon. Like if you're not thinking like a quant and sort of back testing using these tools to make better decisions, like I think you'll be at a big disadvantage. But um but yeah, IR like firms I'm surprised by are actually like you know, build spinning up like a parallel uh a parallel clone of their investment process and actually in some instances actually starting to run small amounts of capital of like you know, this is our human-driven portfolio, but let's do something where the agents make the decisions. And I think that's a very interesting A B A B test.
SPEAKER_00I look, I think this this job is is at least two parts. There's the stock picking and there's the portfolio management. And they're both so integral. Like we all know so many people who were phenomenal stock pickers at the funds that they were at, went and started their own thing and flamed out horribly because risk management, portfolio management is a very different skill set, something that you have to learn on your own too. And so it's funny when I think about it, not having read the article, okay, that you did, or having any context, or really knowing all that much about how Magnetar operates, but I can, I think, probably fill in there and make some generalized assumptions. I'd almost flip it on its on its head. And I would put the AI at the top side doing the portfolio management and risk management and doing the idea gen at sort of the the top of the funnel or the bottom of the process. Because at least in my experience, the out the true alpha, the opportunities are really in that pattern recognition, are really in seeing something that the data is not necessarily showing, or in having that great meeting, or knowing that industry, or some sort of informational edge that really is still very human in nature, and then going up and you know, putting that in a portfolio. And then what the what the AI potentially has the ability to do is adjust factor risk, especially in a long-only uh fashion, think about all kinds of things, currencies, whatever, all regulatory, all that stuff, and do all of that on a more automated fashion at the top level where you're preserving that underlying sort of alpha that's very human at the bottom.
SPEAKER_03I I would I would almost flip it on its head, but we have pressures in that and like buy side alpha capture and like quant center books. Like I think there's sort of like, you know, every quant I've ever talked to is sort of like no, you can't give it to any quant and any human has sizing alpha. And so, like if you sort of like want to adopt that mindset, like here's the raw idea, construct a portfolio, factor neutralize it. You know, I think I I'm sort of on board with that. That I I also think the idea gen pattern recognition can be automated too. So maybe like all of it can be automated. I don't know. We were sort of just lost like meeting with management. I mean eventually, yeah. Um yeah, it's kind of I guess sort of like more scared the more I have these conversations and think through this, but uh someone's gotta go raise someone's gotta go, you know, to Switzerland and raise the capital, I guess. So uh yeah.
SPEAKER_00Yeah, well, maybe AI can do that too. So no one's no one's gonna have a job.
SPEAKER_02Someone someone asked me like what like what skills should I have, you know, a college kid. I was like, be good at golf and be funny. Like two things an LLM can't do and humans love. Yeah.
SPEAKER_00Yeah.
SPEAKER_03Um maybe just sort of closing question. Um, you talked about the importance of AI native, and like part of our mission in all of this is sort of like uh help people chill on their sort of existential dread a little bit, maybe maybe help me chill on my existential dread a little bit. Um what advice would you have for like the younger people to one sort of um uh become AI native? Like any sort of tactics on actually learning this stuff? What have you seen in the interns you've worked at worked with and and and uh and um sort of hired? How can people that want to make a career here in this wonderful craft of investing sort of make sure when they show up on the desk prepare prepared? What resources or methods have you found to be most helpful?
SPEAKER_00Yeah, it's a great question. Look, I think if I think back to my own personal experience, like when I was younger, and what's you know, and I and I'm not uh arrogant enough to say I've been wildly successful, but at least what's made me happy and what I'm doing, I think people should not have the existential dread about AI and attack it by thinking about this as a job skill, as a professional skill. Think about it and becoming AI native as a life skill, because I do really believe that's what it is ultimately. Whether someone's gonna work in finance or not, or they're gonna work in law or law enforcement or whatever they're gonna do, I think in 10 or 20 years, like this younger generation is gonna be using AI in every facet and aspect of their lives. And I the reason I brought up the way you know I grew up because I think, look, all of us on this call, we kind of came up with like AOL Instant Messenger and sort of the early to middays of the internet. And we are internet native. Like, you know, our parents were still blown away when we show up and install apps on their phones and switch the inputs on the TV.
SPEAKER_03But when you're AOL IM, you get this sort of door opening thing. I still think the butterflies, you know?
SPEAKER_00Yeah, exactly. And so that look, that's how we grew up. And we didn't know it at the time, but those types of things that we were doing online, just dicking around, really actually ended up benefiting us because we all went to then go work for older guys or women at hedge funds, and we were doing things, and they're telling us about the days when you had to get a 10K mailed to you, and the proxy came to you in the mail. And so we were it was like never occurred to us that that was like even a thing. We were like internet native by nature. And I think AI is the same way. And so my advice would be first of all, chill with the dread, because you shouldn't view it as a professional thing that you have to do. You should view it as like a life thing and you just sort of relax and let it happen. And related to that is just do it in your free time, the same way we screwed around on the internet and with email and AOL and ICQ and all that other stuff, just kind of mess around with it. The cool thing about AI, which is so different from like the stuff we experimented with, is it actually helps you learn it. So if you don't know something and you're messing around with it, just ask it and it's gonna tell you. And that's really cool. That's thankfully, I mean, helpful for us, you know, now being older about this now too. So my advice would be you know, rather than do whatever people do in their free time now, just like mess around with it. Open-ended, sit with it, mess around with any whether that's you don't have to do claw, you don't have to do codex, sit with Gemini, sit with ChatGPT, and just mess around with these. And it's almost like this, you'll have this, these sort of like artistic moments where you'll realize, oh, I could do because I did this with that yesterday, then open this door, this light bulb went off in my head. I can do something like this. You know, you're gonna get asked to do something at work one day, and it's the light bulb is gonna go off in your head. Oh wow, I could use an LLM to do this, I could use this to do that. And that's just because you would have played with it, you know, before in your free time. So yeah, I I get the existential dread. Um, but I think if you step back and just say, look, I've got lots of time, I've got to play with it. It's still so early days, and this stuff is changing so fast too. That's also why it's inadequate to look at it as a professional skill. You may learn the hell out of it for the next two weeks and like cram it, and it's just gonna change. Um, and so um, yeah, I think just spend as much time as you can with it. Don't force yourself to do it. But whenever you have free time, just mess around. I challenge myself all the time now too, and I love it. Any aspect of my life, if I can sort of outsource it or think about having AI do it, you know, for me now, um, I'll just do it. Like, like I said, we just we just moved in a new home. There's so much stuff to assemble, like all kinds of things. And so I'll just take pictures of instruction manuals, I'll take pictures of like things that are sitting in packages, and I'll ask Gemini, I'll be like, how long do you think this is gonna take me to do? And then I'll go with the one that takes like an estimates it's gonna take the shortest amount of time to do. Just everything you can do in every aspect, just like go with AI the first time and try to figure that out. Um, and I think that'll I think that'll arm people well.
SPEAKER_02I just rewatched seven years of Game of Thrones, seven seasons of Game of Thrones with AI. Oh wow. It's very enjoyable for like a complex show like that to be like, what was really happening in that this character's mind during this moment? Yeah, it's so awesome.
SPEAKER_01It's like made it so that's a really cool way to do it, and it's so layered, right?
SPEAKER_02And like the other thing, too. Something like Game of Thrones has been written about so much. So, like the hive mind knows a lot about game, a lot of opinions about Game of Thrones.
SPEAKER_00Yes, oh god, what the LLMs must think is acceptable to humanity based on what happens in Game of Thrones characters do to each other is terrible.
SPEAKER_03I took your advice on that, Kay, when I went to a Musee d'Orsay in Paris, and I was looking at Van Goghs and like, you know, tell me about this. And it's like, yeah, it's a really way, like, cool way to enhance an existing experience to sort of just go deeper, uh deeper on it. So, yeah, that's a great, you know, yeah, not using AI to bypass an experience, but to actually deepen an experience. Yeah, yeah. This was awesome. Thank you so much, Raj, for for for being with us today. This is really fun to chop it up and uh we'll uh we'll check in in six months, maybe, and we'll sort of look back at the June 2026 episode as like completely anachronistic and everything has changed and uh etc. But uh it's sort of fun to uh fun to sort of build and learn together. So thank you so much for for for being with us.
SPEAKER_02Yeah, thanks Roger.
SPEAKER_03Thanks for leading these great conversations. Awesome. Thanks, Roger.