947: How to Get Hired at Top Firms like Netflix and Spotify, with Jeff Li

22:02Jon Krohn:Maybe you don't really care about explainability so much there because they're seeing lots of different options. It's not like a mission critical decision. whereas some things like predicting maybe where to allocate some financial resources, what market to put advertising money in or something where it's potentially large sums of money and you don't want to be doing that without understanding why you're doing it. Yeah, yeah. So I think you got it right. And if I were to bucket the two things you said, there's two common types of forecasting problems. You have operational use cases and you have like strategic use cases.

22:44So I actually learned this from a forecasting course I took from, uh, his name's like Tim Januszkowski. Um, I don't know if you met him, but he, um, he, uh, basically bucketed this, this concept into like operational forecasts where you take, you build these forecasts to help run your operations. So let's say that you're trying to, let's say Amazon, right? You're trying to predict where deliveries are going so you can kind of effectively run fulfillment. In that scenario, you will probably want the most accurate model. You don't care as much about explainability, but if it's really, really accurate, you will get significantly more efficiency gains from the operations of it.

23:25Versus there's the other bucket called strategic forecasting. And that's typically like, say, FP &A, where they're saying, hey, I think this is where the business is going to be. We need to make decisions off of this forecast. So typically, if your forecast is for like a strategic purpose, you want high interpretability. If you want an operational purpose, you don't care as much about it. So I found that I've worked on both those problems and like with those operational approaches, you can actually use more deep learning, you know, sexy models for those approaches if you can like prove out the accuracy.

23:59But then if you're like, you're trying to like make decisions off of it that are strategic, then you actually really want to know the assumptions going into the forecast.

24:07Jon Krohn:So that was a really elegant, nice way of bifurcating the kind of hand wavy idea that I was describing. Yeah. No, like trigger that. So yeah. Perfect. Yeah. Really cool. All right. So we've talked about some of the approaches that you've been applying at top companies like Spotify, Netflix, DoorDash, some of the most competitive companies to get into for any kind of role, you know, whether it's data science or software engineering or marketing or even probably people working in legal, HR. Those are probably some of the companies that people most want to get into. What kinds of tricks do you have for our listeners to get hired by these top firms?

24:44Yeah, I have so many. So I would say, I'll kind of talk through all the tricks that I used. So when I was trying to get into DoorDash, there was a technique I learned back then called the briefcase technique, which I think actually still works. And I've talked about this on a number of podcasts. but um basically the idea is that like the core principle of getting a job right it's not like interviewing it's actually can you add value and solve that company's problems right so when you an interview is just a way to kind of test your skill set for to see if you can actually solve their problems but in reality you can actually just circumvent this and actually see if you can solve their problems so an example is actually at doordash um like i what i did was i figured out what their biggest pain points were and then i like put together a doc outlining hey these are your problems this is how i would solve it um and i like it was pretty detailed i spent at least like a couple days on it and then you send it to the hiring manager and sometimes it'll hit it's not always going to hit but you have a reasonable shot for it to hit and if it does hit then they'll they'll actually be much more sold on saying hey okay like i'm this person has a skill set that I could solve my problem, I'll bring them in again to have a conversation with them.

26:02So I found that technique to work pretty well for -

26:06Jon Krohn:Do you mind if I interrupt you for one quick sec on this one, on that first point there? Something that's interesting today is, so you started working at DoorDash like more than five years ago, I think, if I remember. Yeah, it was a while ago, yeah. And so back then, obviously, we didn't have generative models that could be creating. So if somebody gets an email from you that's like, these are your pain points, and you clearly spend a lot of time, a couple of days. This is quite an unusual email to get probably for a hiring manager. And that's part of what makes it such a great tool for getting hired.

26:36Jon Krohn:But I wonder if today a hiring manager might say, this is definitely Gen.AI. You know, and it's almost, there's kind of like a, it's almost like the more effort you put into it, the more crisp and perfect it looks, they might say, maybe they're more likely to say this is something that was just created by Gen.AI This person doesn't, like they're just, you know, they've figured out some relatively simple agentic workflow to be like spamming tons of hiring managers with these like, what look like very complex reports. Cause I get that, you know, for the podcast, for example, or for my consulting business for YCaret, I end up getting these like huge PDFs, like 30 page PDFs with illustrations, tons of detail.

27:20Jon Krohn:But I get them like, you know, once a week for each business from some different random person who's just doing a cold outreach. And I'm like, this is definitely some agentic thing going on. Yeah, I see, I see. So yeah, I guess like for, yeah, it's an interesting point because yeah, back then it definitely would have worked. I think today, I still don't think it's, people are doing it as much for like these big companies. Definitely for you, it's like you have an audience so people are gonna reach out. Well, and it's a different kind of thing. Like I'm getting sales, I'm getting sales pitches.

27:52Yeah, I see.

27:53Jon Krohn:And so I think that it could be something very different. Maybe nobody is, I mean, because actually, because I haven't been getting those at all, people reach out and say, are you doing any hiring? And it's a relatively simple message. Nobody has been sending me these big, like these are your pain points. This is how my data science expertise, this relevant experience I have could be useful. So maybe you're right. Maybe it would stand out anyway. Yeah, I think so. So I actually, so I do agree that for the Jenny Eye piece, it's like people, you can easily spam that for sure. I think the hard part is still like you want it to hit like like if you have a Jenny eyes because I think a lot of times even if I do this like if somebody does it to me if they don't if they didn't actually understand my problems then it's it's not really going to hit then I'm going to ignore it but then if they actually said hey like I listened to all your podcasts and I figured out hey your podcast is maybe kind of like you know you could use this kind of software I don't know like you can like improve this aspect of it and they like, and it was like very specific and you get, it was clear that they understood what you needed, then it would hit a lot better.

28:56Jon Krohn:But here's the magic button you press to 10 X your audience and revenue. Exactly. Yeah. I mean, it's like, it's also, so, so I do think that was one trick like that worked back then. I haven't tried it recently. Cause I don't think I need to as much these days because I think as you get more experienced, it's easier to get your foot in the door. Sure. But I would say my recent job at Netflix, the way I got it actually wasn't through applying or using any kind of tactic. It was basically because I had been working in ads at Spotify. I was working in forecasting. So I had this like unique intersection of skills in ads and forecasting at Spotify.

29:38And then when I, when Netflix decided to start doing ads, they needed for forecasting is essential to, to running an ads business. Uh, they needed somebody with that exact experience and that exact overlap and skillset. So then I was like, basically the perfect candidate. And then I, when I interviewed, it was like, you know, pretty, like, I I've been like thinking about this stuff for the last like few years. So it was pretty smooth. But I had been rejected multiple times before at Netflix. You had been. I had been. Yeah. So I think the key thing is, is like you have to just keep trying even if you get rejected.

30:16And then also too, it's like, do you have a unique intersection of skills that is hard for the person to hire for? Because I think there's many people that will do forecasting. There's many people that do ads, but there's a lot fewer people that do both. So I do think that as people like develop their careers, it is good to build some sort of niche expertise in an overlap of say industries and skillsets. And that's actually what really like differentiates yourself from the rest of the market. So I love that.

30:44Jon Krohn:That's a great soundbite. It sounds like that's going to end up in a YouTube short later on. Yeah, sweet, sweet. Yeah, because that's perfect. All right. So switching gears now from finding the perfect job to finding the perfect partner. You have a startup called yourmove.ai and you've scaled that to over 10 ,000 users. And I'm reading a quote here. It promises to perfect your dating profile and put your texting on cruise control. So it helps users sound witty, flirty, or funny on demand, depending on what they're looking for. And so, yeah, it's kind of blurring the line between human expression and machine creation.

31:25Jon Krohn:Tell us about how you got into this particular thing that you're doing yeah yeah so i'll caveat i don't i don't really work on it anymore so it's like a old startup um but i'm happy to talk about it so does it still does it still work yeah it still works it's still yeah it's still it's still going um so my co-founder at the time dimitri he um he he basically uh this is when gpt2 came out he basically wanted to build a like an app that helps you text better and then for me like in 2019 i built a i trained like some deep learning model on like photos that i swiped on and then i would have the deep learning model predict whether i'd like a profile or not and then it'll auto swipe for me um so then i thought my my friend dimitri's project was a natural continuation of what i had built before but i I think I thought I had more legs because I think the dating apps would definitely ban me for trying to like auto swipe on the, on the app.

32:29So yeah, really the idea of your move, um, was to help you write better messages, write better profiles. Also, um, you know, there's a photo service now, so you can actually use AI to help you craft, you know, better dating profile photos. And, um, yeah, really it was like, I was very interested in this space because, because I was trying to figure out dating. And then, yeah, I was like, okay, how can I use my skillset to apply to an area to make my life better? And that's kind of where it came from.

33:06Jon Krohn:It's the, and I mean this in a genuinely affectionate way. This is the nerdiest way to approach dating. I love it. Well, no, everyone's like, they're all like, if you spent all your energy building this system to do it, you could have just done it and gotten the same result. But as kind of data science engineering people, we like to just build systems to do it rather than do it ourselves. And it sounds like it's been effective for enough users that a lot of people have been using the product, have been using yourmove.ai. Was yourmove.ai useful for you? Did it end up having a real-world result for you?

33:44At the time when I was using it more, it was helpful, but it definitely could not replace it cannot i cannot let it be fully automated like i think a lot of the message suggestions it'll give me some ideas but i'll still craft it in my own voice so there was that and i think a couple years ago uh the photo the photo tech was not good enough yet even like today i think it's still like you can kind of tell it's still ai generated and and Like if people see that you have an AI generated photo, it's actually like a negative, like a negative perception. So I still think the photos are not quite there yet where it feels like really genuine.

34:26Yeah. So I would say like it helped, but it could not be the sole thing that like really solved all dating problems.

34:33Jon Krohn:Yeah, that makes a lot of sense. It is interesting that I was reading actually just yesterday at the time of recording that a lot of the big dating apps, which saw surges in their share price in the U.S. at least during the pandemic. So kind of around 2020, Match Group, which owns a bunch of these dating apps, I think like Hinge, Tinder are all owned by this Match Group. Don't quote me on this. I'm doing this from memory. Yeah, yeah. No worries. But then there's also Tinder. and so yeah, Tinder, Grindr for some people that's the perfect dating app and with all of these apps actually they've had a decrease in user activity so kind of post-pandemic with people wanting to meet in person again wanting to have that real connection and so share prices for the publicly listed ones like Match Group have plummeted and apparently all of these companies are betting on AI solutions to improve the app.

35:32Jon Krohn:And I wonder if that's going to work out. So basically these kinds of things, like you did with yourmove.ai, it sounds like companies are integrating more and more. And what clicked me onto that is that, of course, relative to if you were doing, if you got started with yourmove.ai in GPT-2 era, that was like, I mean, there was a 50-50 chance you just got nonsense out of it. And whereas today you can get, obviously, I'm sure everybody who listens to this show uses conversational agents regularly and is kind of familiar with the state of the art. It's mind-blowing how accurate and helpful these results can be.

36:07Jon Krohn:And so integrating that into an app, I can see the idea and it's easy to imagine being in a boardroom, like how are we going to get our share price back up? AI, that's probably happening in a lot of boardrooms all over the world. But I wonder if all of a sudden everyone is doing that in a dating app, like getting help with what photos to select and yeah maybe with photo selection and stuff that's good because why not show yourself in the best light yeah but if everybody's having ai generated messages and ai reading of messages it's kind of a weird environment it feels like to me yeah i mean that's like when we were working when i was working on your movie.ai that was like the biggest concern they're like is this going to be this like dystopian future where every everyone's ai is talking to each other and setting updates for each other um i i personally think now like it's been a couple years since i worked on your move i've noticed a trend to be a little bit more towards in real life so like there there is like a strong especially like being in here in new york i have like a bunch of single friends there's a strong appetite to like go to run clubs go to like in-person events to try to meet people and like there's a really like strong negative perception towards dating apps but people will still use it because that's like the easiest way to like get dates but there's like a strong negative perception so it does seem like it's still going but there's it's like the trend is starting to kind of shift a little bit um i found for me when i was dating being social like doing hobbies that i loved like going out to events that like aligned with my personal values tended to to work better for me but you know you have to use it so like you know it's it's still gonna be a part of what you what you do yeah i think it's uh i think that the the volume that you can get through a dating app yeah the volume of potential people that is it's it's a double edged sword because while theoretically it allows you to see more options that prospective option is also seeing more options yeah and so like you know if it's something like a run club and you two are you know you and the person that you're interested in are both regular runners like you know the number of single people that are regularly going at the same time to that run club like it's going to be end up being a pretty small pond and you're going to feel like hey this is like you know the person for me because yeah you know i keep seeing this person and they're just you know it's just the perfect fit whereas online you're just like well i could see what else there is yeah yeah so so my girlfriend right now um i don't think i would have matched with her on a dating app like we met through like ski friends so we like both love skiing and snowboarding like she was filtering i'm five nine she was filtering for six feet plus guys so like we would never have met on a dating app so it's like you know the medium kind of like forces you to behave in certain ways as well so it is um yeah it's unfortunate part of the game yeah yeah that is interesting that's a really good it's a really good data point solid data point on why uh real life can be better so yeah so interesting things happening in the dating world because of algorithms and now AI, this automation of love.

39:18Jon Krohn:And we'll see how that goes. All right, moving on from the dating conversation, which maybe some of our technical listeners are like, get back to some data and AI stuff more explicitly. So we'll do that right now. So something that you've argued before is that automation with AI fails. Automation with AI fails if people first don't have mastery. And so I'd love you to fill us in more on what you mean by that. So is it that like in an organization, you can't have success just through agents automatically doing things if there isn't some level of human mastery in the system already? Yeah, so I can give a concrete example.

40:05So I tried on the side, right? When AI was like really popping off, I really tried to, I was like, hey, like, you know, there's the gpt image launch i can create ads with this and then if i can promise say bit like founders that i can just grow their business by creating ads and they launch it on meta ad manager then it's like a clear business like opportunity here and i think i still think it's like a good business opportunity but what i realized was like i basically tried to build this ai workflow that would automatically create ads for me and i i was able to create ads but the problem was the ads were horrible.

40:45Like they were not good. I would show them to like a person running a business and they're like, these, these, these are not good. And I didn't have an eye for what was good and what was bad. So like I had never made ads before. I knew how to like hit the open AI APIs. I knew how to like write Python scripts. I knew how to generate the images, but I had never like created ads before and i think because i never had created ads before and i didn't have an understanding of what was good or not like i i wasn't the right person to try to like completely automate this workflow like ideally it's like somebody who also knows ads but also has the tech skills or you know you partner with somebody who knows how to create ads uh ideally that's like the right setup where okay like this person has an intuition on what's good in the system and then I can like automate it, that will tend to work much better.

41:41So I learned from that project that it's better to like know how to do the workflow manually first and know that you can get a good result and then you can automate it. So yeah.

41:52Jon Krohn:It's interesting how there's also how you can carve up any one of these topic areas into different parts. Because earlier in the episode, you were talking about how your expertise with advertising was one of the key things that allowed you to have such a specialized niche and to get hired at Netflix after trying multiple times. But despite all of your experience working with, in the ad industry and working on ads, you didn't have experience with creative, with creating the copy, with creating the image that goes out. Correct, yeah, I didn't have it. I thought that because I had the background in ads that I could do it, but then I never actually, yeah, I didn't have the expertise in creative specifically.

42:32Like I can create agentic workflows doing forecasting and automate time series analysis, I feel very confident in that. But in creating like high-performing creatives, I'm probably not the right person to do that.

42:46Jon Krohn:So my last question I asked, I mentioned agents. We talked about agents a little bit. You have a popular tweet where you condescendingly, condescendingly, I think that's the right word, scathingly certainly is the right word, said that agents are just Python scripts. Yeah, yeah. And so what do you mean by that? Is it really that simple? You know, do you think, yeah, what do you think about agents? Is it overhyped? Are we gonna get, are we gonna see a lot of value out of the money? Because you have mentioned them a few times in this episode already in ways that sound like you see some potential there.

43:18Yeah, yeah, I think I was probably a little too, I think it was a little too condescending and scathing. But I think I'm realizing, right, like, as I try to build, I'm very pro agents. I wanna like build them. I'm trying to get more projects like this. It's like hot. It's fun. But like what I'm realizing is that when I want an agent to automate something, a lot of times like passing it to an LLM, the space in which you can answer is too wide. Like I want a more deterministic, hey, do this exact thing how I want you to do it. And when I really want to make it as deterministic as possible, it then just becomes a Python script.

44:02So like, but I'm very pro agents. I'm pro trying to like automate my job and automate parts of my job. And yeah, so.

44:09Jon Krohn:Yeah, you're hitting the nail on the head there with some of the bigger issues with agents is that, you know, I talked five, 10 minutes ago about how, you know, LLMs are so powerful these days. And obviously that's just going to get better and better. yeah but the flip side of that the double-edged swordness of that is that it can mean a very wide variety of possible responses and moving further away from uh something deterministic something predictable there's different kinds of approaches that people you know you can turn down the temperature on the model to try to have things be a bit more deterministic but but getting you know getting that right can definitely be tricky and then there's also like frameworks out there for testing lots of possible different responses but either way you know i think there's a i think at least what what you're highlighting here is that you know getting an agentic workflow to work effectively in a real world commercial use case especially where it's going to be going at scale is really hard work it's hard and um yeah a lot of times like you wanted to do like especially at like a really large scale a small mistake could be really costly you want it to like work exactly how you want it to work versus like you know maybe there's a five percent chance it might might give us a like a weird answer we want to like reduce that risk so very pro agents i think one of the things i learned recently was actually like it's you can have like the the work execution be much more deterministic but the decision to to trigger that execution actually should be like through an LLM.

45:41So you can have an LLM make that decision on whether to trigger or not. And then when it makes that decision, then you just have like a deterministic workflow to run for that. So I found that to be like a good balance, but we're still seeing.

45:53Jon Krohn:On the topic of agentic AI, something that's very trendy these days is context engineering. And you also have a tweet about how a monolithic code repo can act as a great automatic source of all the contexts that you need to engineer for a particular kind of problem. Do you know what I'm talking about with this? Yeah, yeah, I know what you're talking about. Yeah, yeah, so I think one of the things as I've gotten really deeper into the AI space is that, and this is for me, my personal workflow, if I have 10 different repos scattered across, it's hard for me to say, go into Cursor, have cursor figure out all the files in all these like five to ten different repos and figure out what it's doing and like bring that context into what I'm trying to do versus if everything is in like one place it's much easier for the say cursor or cloud code or codex agent to go and find the relevant files and and start doing things with it and like when everything is in a monolithic repo So then I can just say, hey, like reference this folder, tell like, you know, make a Python script similar to like whatever is in this folder.

47:10And then it'll just like do it how I want it to do it. So yeah. So that's kind of why I was like, I'm like much more pro mono repo now with AI versus before I was actually kind of like against mono repo. So yeah.

47:23Jon Krohn:It seems like the kind of thing that it was a software best practice to say, you know, split out different kinds of functionalities or use cases into different repos. Classic like microservice architectures where everything's separate. And by changing one repo, you can feel like your API isn't gonna be impacted. But yeah, it's interesting how things change now in this world of huge context windows for tools like Cursor. And actually is a great next follow-up question. At the time of recording at least, and these things change quickly, you shared recently that your current So your personal tech stack that you use, you mentioned cursor just there, but you've said that it spans Claude, Whisperflow, Gemini, and V0.

48:09Jon Krohn:So let's dig into those a bit, maybe elaborate on your stack a bit and kind of where these fit together. So for example, Claude, probably a lot of people are aware of is one of the leading conversational agents out there in general. Gemini, similarly, but whisper flow I haven't even heard of and v0 I've heard of and I'm kind of kicking myself that I don't remember off the top of my head what it is sure yeah I can talk through it so yeah whisper flow is basically a voice dictation software where you basically can press a button and it's called whisper flow because you can like whisper into the mic and then it'll actually dictate it pretty accurately because you're like in an office you don't want to be like talking super loud.

48:54And then I found that to be really useful to like leverage with AI because it's much easier to like say everything on my mind than to like type it, type it all out so I can actually inject a lot more context when I'm trying, when I'm prompting, say the agent or the AI.

49:11Jon Krohn:So you'll be kind of sitting at your desk and you'll just be whispering into your mic on your laptop. Yeah. Yeah. Or if I'm working from home, my, my girlfriend will just hear me just like saying a bunch of stuff. And then she'll be like, oh, he's talking to the AI right now. So yeah, yeah. Nice. We'll see. Our listeners probably just heard a beep on that first version of stuff. Yeah, yeah, yeah. They can use their imaginations. Yeah, it's okay. You can swear. We'll just bleep it. Okay, cool, cool. And by the way, listeners, I probably haven't mentioned this in a very long time, but the reason why we bleep, because I'm actually, I swear pretty liberally in my day-to-day life.

49:47Jon Krohn:but if you swear one time in any of your episodes, you have to change, you have to tick this box across all podcasting platforms. You go from like a suitable for everyone podcast to like an adult's podcast. Oh, interesting. Isn't that crazy? Yeah, it's crazy. So yeah, so we bleep and then that's the way around. It's pretty funny. A few years ago, it was our previous podcast manager. She's Eastern European. um originally and she kind of she sent me this list of swear words and was like which of these words do you need to bleep out and which ones can we leave in that was a fun one um i'd go into it but you just hear a bunch of bleeps uh on the show so whisper flow that's cool i like the idea of that do you think that there's a difference when you when you take the time to write an email So now with my consulting business, one thing that I'm trying to make sure I do is convey things clearly to a client of my consulting firm and to make sure that their blood pressure is staying low on kind of everything that we've been thinking across the details of their project.

51:01Jon Krohn:And I think if I tried to dictate that, I might not be able to do a good job. Although, even as I say that, it would be a great starting point maybe for the email. You know, I could kind of, I could stream of consciousness. And even if that stream of consciousness isn't very clear, I don't need to go from my stream of consciousness to transcript. I can pass that through an LLM on the way and say, this is going to be an email to a client. You know, can you take this audio or this transcript and convert it into a nice structure? And then I could review that. So actually I've kind of, I was going to say, do you think there's a downside to just dictating, but I've talked myself out of it?

51:43Yeah, I actually would say like for prompting, I dictate a lot. But if I'm trying to like write my own, say doc or like a spec, I actually just write it because I find that like writing it helps me think through things. So like I, for a lot of cases, I actually don't dictate and I, I don't dictate and prompt. I just actually will write it. So I actually properly think through it. But if I don't think I need to think through it, then yeah, I'll probably dictate. So Nice.

52:09Jon Krohn:Okay. So we do kind of agree that there would be situations where dictating might not be the best place to start. And so you're saying that, so it's when you're prompting that you're often dictating. So it's, so you're literally, so you're talking to a conversational agent, you're talking to Claude or Gemini. Yeah. Yeah. If it's like not too disturbing to people around me, then yeah, I'll try to. I'm going to have to try it. I've literally never done it. Yeah. I think if you use cursor now, they added a dictation feature. So you can, you don't have to like download, you can just try the dictation feature in cursor and see, see how well it works for you.

52:38Jon Krohn:Nice. I like that. All right. So yeah, so let's talk about some more of these tools. Why do you use Claude, Gemini, and Cursor? And what is v0? Is that also kind of of the same ilk? Yeah. So, okay. So I use Claude code for the terminal agent. So if I'm running the agent through the terminal, I found Claude code to work better for me. I haven't tried codex. I know that's also a popular one that's like growing, but I've just found Claude code to have the most number of features like they have some cool features like skills they have an agents feature where you can actually like save your agents i've liked those um but recently a cursor just released their new model composer one which is crazy fast and i've actually started to like move over to using that model a lot more so you know this stuff changes you know week to week um gemini i will use actually more for image generation i found nano banana to be like a lot better than the chat gpt image generator so if i want images i'll use uh nano banana and i test this with um image like i try to test different hairstyles using my face and chat gpt will like mess up my face versus nano banana will actually maintain my facial structure and then actually add a different hairstyle so i use that kind of to test the image generation to see how well it captures my keeps my face is that it Is that because did you stumble upon this particular test, this particular validation mechanism through a real world use case?

54:10Jon Krohn:Yeah. Yeah. I was like growing my hair out. So I was like, oh, I kind of want to know what I look like with longer hair. And then I did it with chat GPT. It was horrible. I was like, this doesn't look anything like me. And then I tested it with Nano Banana and actually like looked pretty reasonable. So I was like, oh, yeah. like i think like maintaining the facial structure and making it like look like you as a person is still hard um so yeah so i found nano banana to just be much better yeah i recently had the same experience with uh i had to update a social media uh card that for a podcast episode a super designs podcast episode that i was posting because i gave the artist who creates our thumbnails the wrong long title of the episode.

54:53Jon Krohn:I had accidentally like in the file I left in a previous episode title. And so, but I didn't notice that mistake until it was time for me to post it. And so I went to chat GPT first, just like you did, because we often, I think we often think of, and cause this was the case for a long time or for at least a year or two, where across any kind of, across almost any kind of generative capability, you know, there were points where maybe you'd say mid journey for some kinds of image generation tasks, but a couple of years ago, it was like OpenAI is the place to go. Just start with the OpenAI tool because you're probably going to get the best results.

55:28Jon Krohn:And so that's where I started, kind of out of habit. And I couldn't get the image to stay, my face and the guest's face and the whole layout of the image to stay exactly the same. And it kept doing the exact same misspelling of a word. It was the word accelerate and it was missing one of the E's. and it just could not get it to work. Whereas in Gemini, oh, you know what? I think it was the other way around. I mean, it just goes to show how quickly things change. Because now that I'm actually saying this out loud, I think it actually, yeah, it was in Gemini that I had the frustrating experience where I just couldn't get it to spell this word accelerate correctly.

56:09Jon Krohn:And it couldn't even seem to, it's amazing how you can get into these loops with the agents where I was like, it just could not, I'm just like start from scratch and just do it over. And it was just like, I actually, it just said, I can't do this. I'm like, what? Like, how can you not do that? You can definitely do that. And then with chat GPT, I think the reason why I'm remembering this is because it gave me multiple outputs. And it was like the first one I clicked on, I was like, oh, it doesn't look good. The second one I was like, oh, it doesn't look good. It changed my, it changed our faces.

56:38Jon Krohn:And then somehow the third one was just perfect. So yes, it goes to show the stochastic, kind of tying back to your point about determinism or even something like that, where like, obviously we want the image to stay the same in this kind of use case. And two times out of three, it gets it wrong by luck. I guess one time out of three, I get it. Yeah. Yeah. I remember when I was doing the ad creative thing, I like, I think I spent like$200 of API credits to figure out how to like prompt it in a way that would get it like exactly right. But it took like a lot of experimentation to like figure that out for me at least.

57:13Maybe it's like much better now. So yeah.

57:15Jon Krohn:that can end up being ip for some people in uh some use cases you know just getting that prompt right spending a lot of time on getting things uh just right for that particular model and then they do a model update and yeah then it changes so you always have to keep playing with it yeah um yeah i could talk about the other tools if you want as well oh yeah for sure yeah yeah so the last one is v0 so i found v0 to be really good for spinning up uh web apps so like if you want a front end And I mean, there's like lovable bolts, but I found VZ, I like V0 the most because V0 was originally like software for deploying infrastructure.

57:53So I've found that they have pretty good front ends. I can screenshot something, paste it in there. It'll generate the front end. And I like have deep trust in their backend because I've like used V0 for app deployment in the past. So yeah, that's like for spinning up easy web apps.

58:12Jon Krohn:That's cool. I like that a lot. Yeah. And so I guess like as a, you know, as a kind of underlying point across all of this, it seems like just like for me that it's worthwhile having Cloud Code, Cursor, Gemini, V0, and to be experimenting with them for different kinds of use cases because every month, which one is going to be the ideal choice for you for a given and use case changes. Yeah, it changes all the time. So yeah, I think the core of it is really just to keep trying stuff. When new things get launched, just try it, see how you like it. And yeah, I mean, like if one provider changes something that makes things a lot better, you just switch over to it, so.

58:56Jon Krohn:All right, as a kind of final topic area for you, before we get into the final questions that I ask everyone, as a final kind of topic area, you have spent a lot of time in your career, I guess, in your personal life as well, you know, optimizing and also, you know, developing mental models of how to make the best decisions in the world. And so what kind of advice do you have some kind of like, you know, general takeaways from that experience that you can share with us? Yeah, yeah. So I think like I kind of like use mental models and life principles kind of in conjunction with each other. Like I think the mental models concept was inspired by the like reading about Charlie Munger and the way he approached investing was to like read from a wide variety of, of like topics to kind of come pull these like essential principles out of life to apply to like their decision making and investing.

1:00:04Jon Krohn:Charlie Munger there being the right hand man at Berkshire Hathaway. Exactly. And I think he lived in 98 and passed away just recently. Yeah, he passed away recently. So, but yeah, he's been a, he's a huge influence in the investing world. And for me, it's, I found it to be really useful just like to help guide day-to-day life decisions, how I approach life when things do come up, how to like approach things. And I'm still kind of iterating on it. like um like what one that comes to mind is uh this concept this like model of asymmetric upside so when there are like things that occur in your life and there's asymmetric upside you do them like an example is like actually going to parties with interesting people like you know sometimes i might be a little bit like feeling introverted a little bit anxious about like not knowing anyone there but i always found that it's never there's no bad downside to going though there's only upside you could like meet somebody really cool that you want to be like really good friends with you meet a great business connection versus the downside is like maybe you didn't meet anybody interesting you just go home and and then just like move on with your day so like i found that mental model to be really useful another one that comes to mind is activation energy so like in chemistry like first i don't i don't remember the exact terms but like in chemistry you need to hit a certain level of activation energy to have some sort of like a chemical reaction i found that like when i'm feeling kind of sluggish i don't really want to do stuff um as long as i get myself to take some action i know that i will hit some threshold for activation and i'll i'll like get a lot more excited to do something.

1:01:50So that's like another one as well. But yeah, I would say like, it's just kind of a useful way to guide your life and guide your decision-making and how you approach things.

1:02:01Jon Krohn:Nice. I love that. Thank you for those. And so, yeah, now moving along to my kind of my standard ending questions, maybe it'll flow nicely. You know, the topic that you just had there, your mental models will flow nicely, maybe into your book choice for us. Do you have a book recommendation for us? For me, the one that comes to mind is Unleash the Power Within by Tony Robbins. So I'm a big personal development guy. I found that book to be very impactful for me because it really helped me clarify what I valued in my life, like what is important and like what is maybe not working for me. it also helped me really like learn how to instill a high level of belief in yourself and I think you know you need that to kind of be confident and like tackle any kind of goal that you want so I found that like if if I were to recommend only like one self-help book I'd probably recommend that book that was like pretty impactful for me nice very cool he is a kind of a classic name if not the most canonical name in self-help yeah so that is cool have you have you seen Tony Robbins live.

1:03:08Jon Krohn:That's supposed to be quite a show. I did the unleash the power within like a few years ago. Um, it was great. I mean, it's, it's kind of like, like a rave slash like motivational, motivational event, um, at the same time. Um, so yeah, it was, it's like really brings your energies up. It really like, it gets pretty emotional. Like he does some interventions where he like talks to like some people come in with certain problems and he really kind of tries to help them and it gets pretty emotional. I did it a while back, but I found it to be pretty useful. That's really cool. You know, in my mind, I kind of had this impression in my head that, well, you know, that kind of experience, that kind of Tony Robbins thing, that's not for me.

1:03:53Jon Krohn:I'm not the kind of person that does that. So it's interesting to have someone like you who actually is a lot like me in a lot of ways. Say that you got so much value from it, I guess it's something I should be considering. Certainly, I mean, the book is a pretty safe place to start. Yeah, I think read the book if you like it. It helps. It helps when the event is like much more like you'll definitely feel something versus the book. You know, some people have read through it and they're like, oh, yeah, it's like I got kind of bored or, you know, so it hits people differently. Nice. All right. Well, hopefully an interesting wreck, not a boring wreck for most of the listeners who checked that out.

1:04:29Jon Krohn:Yeah, yeah. Nice. And so, Jeff, my final question for you is where can people follow you for more of your thoughts? It's been a really interesting episode where, you know, I've mentioned things like your Data 365 courses. I'll be sure to have a link to those in the show notes. We've talked about some popular tweets that you've had. Where should people be following you? Yeah. I mean, my personal website, jeffleechronicles.com. And I'm on, I'm kind of lightly on Twitter and not like, you know, I kind of just post whenever I feel like it. Twitter is Jeff M as in Mary, Jeff M L I, Jeff M Lee. And yeah, that's pretty much it.

1:05:07Yeah. That's where you can find me.

1:05:10Jon Krohn:Fantastic. We'll be sure to have links to all of those in the show notes. Jeff, thank you so much for taking the time. And yeah, hopefully we can check in again in a few years and see how your journey is coming along. Yeah, cool. Thanks for having me. What a knowledgeable guest. I hope you learned as much in today's episode as I did in it. Jeff Lee covered strategies and tools for effective forecasting and time series analysis in general, his top tips for getting hired in technical roles at top firms like Netflix and Spotify, the promise and peril of ever-increasing automation and AI around finding your romantic partner, and the collection of AI tools that he uses as part of his daily workflow, including Whisperflow, Cloud Code, Gemini, V0, and Cursor.

1:05:53Jon Krohn:As always, you can get all the show notes, including the transcript for this episode, the video recording, any materials mentioned on the show, the URLs for Jeff's social media profiles, as well as my own social media profiles at superdatascience.com slash 947. Thanks, of course, to everyone on the Super Data Science podcast team, our podcast manager, Sonja Breivich, media editor, Mario Pombo, partnerships manager, Natalie Jaisky, researcher, Serge Macisse, writer, Dr. Zara Karche, and our founder, Kirill Arimenko. Thanks to all of them for producing another super episode for us today. For enabling that super team to create this free podcast for you, we are deeply grateful to our sponsors.

1:06:31Jon Krohn:You can support the show by checking out our sponsors links, which are in the show notes. And if you'd ever like to sponsor the show yourself, you can find out how at johnkrone.com slash podcast. Otherwise, just keep on listening. That's the most important thing to me. But share the episode with folks who might like to hear it. Review the episode on your favorite podcasting app or YouTube. Subscribe. but most importantly, yeah, I just am so glad to have you listening and I hope I can continue to make episodes you love for years and years to come. Until next time, keep on rocking it out there and I'm looking forward to enjoying another round of the Super Data Science Podcast with you very soon.