Sean Tibor: Hello, and welcome to Teaching Python. This is episode 155. Hello, World is Dead. My name is Sean Tibor. I'm a coder who teaches, and my Kelly Schuster-Paredes: name's Kelly Schuster Paredes, and I'm a teacher who codes. Sean Tibor: We are joined this week with an old friend, Julian Sequeira, technologist, pythonista, uh, data center, cabling rat in a past life, calendar guru. Calendar guru, yes, man, uh, of many talents. Welcome Julian back to the show. We're super excited to have you. Julian Sequeira: Thanks, Sean. Thanks, Kel. Happy to be back. It's been a couple of months, but, yes, the reason I've been back is I didn't have a calendar invite from Kelly, but I figured out how to get back here, so it's nice to be here. Thank you. Sean Tibor: Well, we're excited to have you and friends of the show and people who have been listening for a long time. We'll recognize Julian's voice. He's been a guest with us, uh, many, many times. And one of the first people we met in the python community way back in 2018, 2019, when we were first launching the podcast. It's always good to be back with friends. Julian Sequeira: No, thanks for having me. And I have pictures of that. I saw them the other day. I was reminiscing on Pycon with Anthony Shaw. Kelly Schuster-Paredes: I love that picture of all of us. It was like 12 of us and one girl. We're gonna talk about that at one later. Yeah, I have. Julian Sequeira: That was from two years ago. Kelly Schuster-Paredes: I get to go with my win of the week. Go ahead. Sean Tibor: Yep. It'll be good. So let's. Let's do that. Let's start with the wins of the week. Kelly, you're excited, you go first. Kelly Schuster-Paredes: Okay, cool. So, speaking of friends, at Pycon, I am bringing a new. I'm, um, bringing. Actually, we have two colleagues. Sean didn't even know this. Bringing two colleagues with me from Pine Crest to Pycon. Um, one of which is a, uh, person that Sean knows, that's Chris wants, uh, Chris. And he's coming with us. And then I'm really excited about. I have a girl coming with me. Her name is Kayla, and she is an aspiring data scientist, data analyst who knows rich. And so I said, what to learn Python. If you can code anything else, then you're going to be fine. And she's like, I'm, um, coming to Pycon. I'm like, yes, you'll probably pick it up faster than I did. So I'm super excited. She's going to be Joining us and they're going to come to the education summit. So that is a super win that just happened two days ago and for once I'm not going to be just going with Sean. Sean Tibor: Yeah, yeah. Get to bring some fresh blood to it. I like that. Kelly Schuster-Paredes: Yes. Julian Sequeira: People, people with personality. Kelly Schuster-Paredes: Kill. Julian Sequeira: That's amazing. Kelly Schuster-Paredes: Crazy. Sean Tibor: I'm sorry, no comment. Julian, how about you? Julian Sequeira: No, that's brilliant. My win actually doesn't have to do anything with code. I'm actually just really proud. So I've got a 10 year old son and he plays basketball and for a few seasons now and the season over here for this little comp he's in goes for one quarter of the year at a time and I think they're in their third or fourth season and he for the very first time scored his first basket and he has shown up every week. He's hated it sometimes he's loved it. He's been disappointed, he's been sad, happy at even just taking a shot. But he keeps showing up and uh, I was just really proud. It's something I just can't shake. It happened a week ago today but it, it was just this moment of pure joy where the ball he shot, it was a perfect shot, nothing but net. And it's like everything was slow mo. Everyone just jumped up and was screaming and cheering. All his mates on the team were high fiving him and I'm pretty sure the other team thought that, you know, the timer just went off or something and we'd won the game because there was just such an uproar from everyone, the coaches jumping around and I'm jumping out of my chair trying not to run out on the court and give him a hug and it was just, it was wicked. That's my win. My son's persistence to. Kelly Schuster-Paredes: Was it like a massive endorphin now he wants to play a lot more. Was it like that endorphin rush? Julian Sequeira: No. Now he said I'm done. He said I'm done. Dad. Yeah, I've peaked. No, he's coming. We got a game this afternoon or tomorrow afternoon so he's ready to go for the next one. And I said the next one gets easier. So um, now that he knows he can do it. Yeah, it was good. It was great. Great day. Kelly Schuster-Paredes: Look forward to. And Sean Goodwins. Sean Tibor: Um, so I had a moment. Was it last night that was just eye opening? Not quite as good as first basket ever but, but it was pretty close. I mean it wasn't anything like that. I have a 12 year old daughter who is in seventh grade and she needed to study for her science exam that she has, I think tomorrow. And it's all about water quality. Kel, you'd love this one because it's all nerdy bio stuff like water quality index, turbidity, color clarity, acidity or ph levels, all of that stuff is in there. But they gave her an 11 page PDF document for this unit. She's like, dad, you know, teacher's really serious about this. We have to use the words that she's using. We have to do the things the way she's doing it. So I'm like, okay. So I want to be very mindful, not just as a dad, an example for my daughter and being in, in alignment with our, the school's code of conduct when it comes to AI and all of those things. I truly want this to be something that enhances learning rather than avoiding it. So the first thing I do is drag the PDF into Claude and I'm like, make me a study guide. I gave it a prompt and it needs to have these parameters. It need a two day study plan and make it something that's engaging and helps to enhance the learning and provide concepts and everything. Guys like, oh my God. It came up with diagrams, a concept map. It showed how the water quality index has all these nine factors, how it's all linked together, pulled all of this from the document. You could click on the diagram and it would use that as another prompt to generate information about ph levels and what it means. It came up with a ph level bar across the screen from 0 to 14 and you could click on any one of them and it would give you examples of ph levels that were in that range. And I'm looking at this going, oh my God, how is this generating all of this information, all of this content? And it was good. Like it was the way I would want to learn it because it's got concepts in there, it helps with vocabulary, it's helping with, to explain connections. It came up with chains of events to say, okay, if you can explain this sequence of events, like runoff from fertilizer leads to algae blooms which then depletes the oxygen levels. So you can connect all these things together that'll help you be ready for any questions along the side. It just kept going and you could drill down as much as you want into it. Like you could just keep asking it more and more questions and ask it for sources. I was completely blown away by how well it worked and what's possible for people, for kids learning now Especially on a subject that's seventh grade. It's pretty standard. It's something that we've probably taught in the most boring way possible for many years, except for when really good teachers spend the time to actually try to make it interesting and engaging. And now you can do that from your laptop and get something that's really cool. Kelly Schuster-Paredes: Now I know why you didn't sleep but four hours he was playing. We found out the reason he was playing with AI and learning about water quality from seventh grade. Sean Tibor: I mean, it's not a bad reason to be up all night. Been up all night with the flu. I'd rather be learning about water quality. Kelly Schuster-Paredes: No, it's really cool. And then now you know they have Notebook LM that does nice things as well and that has the podcast and stuff that they can use for their own. There's a lot of tools out there. That's one of the side jobs that I love. My new side job at school is doing this stuff with all, uh, the teachers and hopefully more and more teachers will be doing that and showing the kids how to do it themselves versus doing it for them. That's the kid, like teaching them and letting them be in there doing it. Sean Tibor: How does it enhance the learning? How does it make the learning better and richer and more engaging? Kelly Schuster-Paredes: Except for they still do not give good metaphors about APIs. It's always going to be the restaurant API metaphor and I hate it over the restaurant metaphor for APIs. Sean Tibor: Well, have you told it not to use the restaurant metaphor? Kelly Schuster-Paredes: And then totally. And then the kids tell it to make like Sephora API metaphors or some sort of other. I don't know, someone did a, like a series they were watching and they were able to explain it. So yes, I just tell them don't give it, tell them not to give it restaurant API metaphor. And it pretty crazy, but cool. Sean Tibor: All right, well, let's jump right into the topic because I think it connects here. So the bold provocative statement, Kelly, is that hello world is dead. Kelly Schuster-Paredes: Sure, Definitely. It's a new way of teaching. I didn't even know what being a non computer science major person, you had to. I remember the first time you explained to me the point of it. So I'm going to let you explain to everybody else the point of the hello world. Sean Tibor: Okay, So I have to do the back in my day sort of thing. Kelly Schuster-Paredes: Yes, the boring, the boring time, the history. Sean Tibor: I mean, I, uh, guess it's become almost tradition at this point that like whenever you're learning a new programming language, you have to get whatever the interpreter or compiler or whatever it is set up, and the first thing you want it to do is emit hello World just to prove that you can do it, that you can get everything set up. And I think the first time I did it was probably like on an Apple IIe running Apple soft Basic, doing it that way on a green screen and hello, World comes out. The interesting thing, or the reason why I got hooked on it was like, oh, my God, I made the computer do something myself. I told it to do something and it did it. And then you went into the next step. What else can I make it do? I know when we were teaching Python before chatgpt and everything else came out like, we viewed that as a very important step. It's your first moment of agency in learning computer science. I'm the one in control. I told the computer to do something and then it did it. The only real difference between back when I was first doing it in the 80s and in the 2000s and 2010s and 2000s was that now it seems like there's a lot more things to set up just to get to the hello, World part. So I'm going to get my Python interpreter or I have to get moocode or something installed before I can even get to that point. Back in the day was like, flip it on and a prompt comes up. Kelly Schuster-Paredes: Crazy. I always felt like, though, there were so many things, like you said, of trying to get started with coding and I didn't have the green monitor. I don't remember what I. Maybe I was doing DOS or something for a limited time. I have no clue because I didn't get hooked on it until actually, what, 2018, when I started doing coding. But the hello World was like, oh, okay. And I always felt that teaching that to sixth graders was like, hello, World. But if I said, hello, Johnny, hello, Susie, they're like, I don't know why. So, uh, just like, that was my take on it. But it seems like a lot of work to get it started and just to get it to say exactly what you typed in. Sean Tibor: Yeah. So it always has been the most important first step, only because it gets you to the second step, which is that you can then change it and make it whatever you want. Like, just getting it to say hello, World by itself was kind of cool. But getting it to say hello, Julian, or hello, Sean, or hello, Stupid Face, that was the cool thing about it because you had that freedom in the agency. And now we need more, uh, just Julian Sequeira: to add to That I feel like there's the dopamine hit of having it do that. And when we were younger, now when I started coding, we were using hammer and chisel on concrete. Sean Tibor: So your own. Your own granite punch cards, right? Julian Sequeira: Exactly, Exactly. There's a dinosaur looking over my shoulder. Back then when I was learning to code and my first hello World was C. And I remember that thrill of going, oh, my gosh. Like, it just exactly what you said. But I didn't have anything else in my life that I had access to that could even remotely do that kind of thing. Whereas now I'm looking at not just the environment, but the people who are coming to code. They're so exposed to it in other forms. They're so exposed to technology, to apps and tools. And you open your phone and it says, hey, Julian, how's your day going? You know, so that kind of dopamine hit thrill of making having a computer almost feel like it's talking to you. That's how it felt when I read my first print statement, is that it felt like, oh, the computer knows me. Ooh, I could ask it a question. What's two plus two? It's for. That's really cool. But now that stuff is just there, so I feel like there's an interest level to this that that makes people go, okay, so what? Why? Sean Tibor: I feel like there's like an inspirational poster here. Like, you'll never remember what your first program did, but you will remember how it made you feel. Like, I think that's. Julian Sequeira: I think that's what we're getting for us. Done, done, done. Um, train mug, quick copyright. Kelly Schuster-Paredes: Yeah, but that. That's. It's funny, though. Cause I'm sitting there thinking about my sixth graders because they're the newbies. They're the first ones that have the taste of thrill. And yeah, it was okay when we did hello World, but it wasn't cool until we made a little app using a little variable and putting their names in. And realiz can do the input with a variable and get a different sentence out. I don't. That third day of coding when we were doing this, it wasn't until that third day when they started to say, oh, this is really, really, really cool. I don't get that anymore. Uh, really? From the 8th graders, because they've already done that. So you have to do something spectacular. And that's why. That's why when we do, like, a web browser opening or a picture loading or an API calling up a website, when the kids start to say, okay, I can see this is quite useful because now I can get into OpenAI with my APIs or something. Sean Tibor: It's also a good point. I'm biased. I've built my entire life around technology and engineering and everything, so of course I'm going to overly romanticize that first moment, that first step into the world. But you're right, for most people, it's at least from what we saw, it took more. It did take more. Now the question is, are they becoming immune to that dopamine hit? Because they can go into a ChatGPT or a Claude or something like that and say, make me an entire website. It just goes and it's all there. So how do you have a dopamine hit of hand coding yourself if in the same amount of time and investment you can have a fully blown out application? And also, is there any value in hand coding it yourself anymore? Are we just really becoming prompt monkeys, working our way through building stuff using agents? Kelly Schuster-Paredes: I mean, I've been having this whole philosophical thing. I finally broke down and installed cloud code in my VS code and I'm like doing stuff, chatting, and it says, just, okay, done. And I was like, oh, okay, oh, okay. I said, oh, uh, don't I need a virtual environment? And I'm like, oh, there it is. Great. Uh, I was not excited about it. I don't know, you would think that me person that's not a computer science geek from the start, a person that's really into AI would be excited about the fact that I got something running with cloud code and it wasn't as. What was more fun is when I told cloud, stop typing for me and just let me cut and copy and paste it into terminals so I can at least feel like I'm doing something. It was like, okay, if you want, inactivate my virtual environment. I was like, stop doing stuff for me. So, I don't know, it's a, uh, tough. It's a tough call. Julian Sequeira: Yeah. That's one of the conversations I was having with a teacher contact of yours, Kelly, over in Prague. One of the things he said to me is one of the challenges you have and that we'll all face is I don't know if we're deviating from the topic too much, but I'll just say this much based on what you said is the why we at least learned with some level of friction to write the code and write hello world and go, how do I go from just printing to now taking input and from input to making that repetitive and catching errors and Whatever. We played with that and we saw that satisfaction of seeing it move forward when you grow up in this environment where it can just all be done for you and likely will be done for you in another generation or so. Getting to this point of why should I know that? And what. Why do I need to know that? Why should I run, as you just said, why should I run the code myself in my own environment? You'd lose market share if you're at. People do that. So it is a tricky spot. And that's why. Back to that topic of hello, world is dead. I just. You're right. Even teaching my own kids and showing them because they go, oh, dad, you're the python guy. So at first, by the way, I'll share this anecdote. At first my kids thought it was super lame coding, whatever. And then they get to start high school and stuff and they start hearing the word Python. Oh, my dad knows Python. My dad runs a Python company, all this kind of stuff. And then now that he's learning it in the classroom, he's like, oh, this is my dad's website. And have a look at this. You know, the Py Bytes platform. When I went to sit there and show him coding the way I learned it, or at the very least through the newbie exercises even that he was just like, but can I just do Minecraft coding? And I'm fine. Then back to that point, you need that extra dopamine hit of like, interest and connecting it to what's meaningful to them. And there isn't. That kind of hardcore desire can rule the world. If I can spawn a whole bunch Kelly Schuster-Paredes: of chickens, that's Sean's favorite thing to do in Python. Sean Tibor: That, that in a, uh, literal megaton Kelly Schuster-Paredes: of TNT for loop of 100 chickens falling down from the sky. Julian Sequeira: That's pretty cool. Okay, I take it back. Kelly Schuster-Paredes: Hello. Hello, chicken spawn. Sean Tibor: I also think you're hitting on something there that's important because no one's going to learn coding the way we learn coding. Like, it's never happening anyway, right? Um, and at the same time, there's value in being an expert and being knowledgeable about how things actually fit together. Like, you can write better prompts and you can guide the process in a more focused, expert way when you know what's going on under the covers. So I can directly the AI agent to make better programs for me, and I can give it specific constraints and I can give it the right keywords that put it in a specific direction. That is a better outcome. I have specifications and Requirements, even if they're in my head and I'm iterating through them. What I don't know is like, I got to that point by making a ton of mistakes and having all that desirable difficulty and the struggle of learning something and having to, um, appreciate the friction in order to get the payoff and the reward. And that's how I got to where I am. What I don't know is how someone starting off coding now is going to get to that same level of expertise, or if it's a different flavor of it, or if it's even necessary, or if it all just becomes this, like, elite, tiny ruling class of AI, uh, experts and coding people who, um, really dictate how the rest of the world is going to work because it's all the agents that they are creating and how they're directing it. Are we even seeing the last generation of kind of deeply. And I don't mean this as a positive, like deeply knowledgeable programmers versus expert prompters and orchestrators of the systems and tools that we need to use to get things done. Kelly Schuster-Paredes: You. I've been playing with things that get a dopamine hit with the kids, using them as my test dummies. That's what I was trying to avoid saying, but I'm just gonna say it. And I'm finding out at least there's three major types of kids, right? There's the kids that will just go out and can create really cool stuff because they've done it with AI and they're still going out and saying, look what I created the one, one type of student that really is going to get into AI. And I had, for example, this kid shows me, do you work in terminal? A little bit. I was like, uh, you know, I know how to activate some things. And he goes, look what I did. And there's like color letters just forever for looping through his terminal. And he says, I found that and I got AI to make that for me. I was like, do you even know what you typed into your computer that you could have done something? He's like, I know. You know, it could crash, blah, blah, blah, security, whatever, it's fine. And so I have that. Julian Sequeira: Who's this child? Kelly Schuster-Paredes: He's a crazy kid. Uh, he did it. And then I have the kids that are really excited when we print out and we do something with AI. We get the code and we go through the code line by line, and it's that moment when they realize one, that AI wrote a function that never got called, and it's never in there. And so their code's not really. And they're like, oh, uh, that makes sense. They get like this excitement. Oh, well, what if I asked it to do this? And I was like, yeah, so that's why you use the words that we're talking about. Use this function. Or that's why you can see if there's a defining function and then a calling of the function later on. And I see that's some sort of dopamine hit that I get from those people that can read the AI code or when I tell them to make it simpler. And then I just have the kids that are old school. Old school. These are your CS people. These are the Shawn Tibers of the world. This kid's like, I don't want to use AI I just want to code this by myself. I was like, you can 100% code this by yourself. I will give you extra time. It's not due in 30 minutes like everybody else. Turn it in when you're done. I don't want to say to you that you have to do it with AI if you want to do it with AI afterwards. So you can do like, crazier stuff or you want to refactor your code, go for it. But it's not mandatory. You have those type of people still that still track to the people that they always were, right? The people that were excited about writing hello world, the people that were excited about trying to break the game and the people that were trying to cheat off of their friends just surpassed the class. It still tracks. It's just a different level of the dopamine hits and it's trying to find where it fits with that child and just get him hooked some way. That's our new world. Julian Sequeira: Like that. And that point of. There's always a different kind of learner. It takes me to. There's an analogy constantly thinking about. And I'm sure everyone has heard this at some point, but it's. It feels very much like a car mechanics, right? I know enough about my car, but if a light turns on in it or there's a rattling sound, I've got no idea how to fix that. I'm that person who's going to go to the expert, but that expert, that mechanic, they're like those CS students who have just spent years honing their skills and figuring out every nook and cranny of that could where a problem could hide. And they know it in and out. They're the ones who just believe it, live it, breathe it, and uh, everything. So they're the people, but not everyone's going to be. Kelly Schuster-Paredes: The funny thing is, anymore, they just plug it in and let the AI tell them what's wrong with the car. Go ahead. Julian Sequeira: That's true. But I made two over the years. They've just looked at my car and said, jules, I'm going to. Just give me the keys. I'm going to fix this for you. I'm like, what's wrong with it? So I'm the kid in that instance who just runs with it. And so someone else build it for me, figure out how it works. I'll run it. But yeah, it does make me wonder if that's where the world is going with coders and developers. And there will be people who dive deep and go straight into it and know the nuts and bolts so that when AI screws up, like you were saying, Kelly, in those instances, writes code that's not used. Or doing some vibe coding recently, and I looked at the code that it spat out, I'm like, that should just be a function. You've repeated that same line of code at least six times through this code base. Like, this should be refactored into a function and just called when it's necessary. Sean Tibor: Something to those personality types, right? Traits that people have and what they bring to it. There's the people who just want it to work. Like, just give me the car and the keys and I'll remember to put gas in it. Most of the time I just want to get. I just want to drive it. The people who want to modify, um, it. I want to have a different paint job. I want to fix this thing or change this thing out about it, right? And then there's like, the people who deeply want to know how it works. They want to understand how every part fits together, how all the pieces go. And I think what all of them have in common with this, with these capabilities, is that pace is happening faster. It's even more iterative. Honestly don't know if it's a good thing, but I think it's a thing. It's coming at the same time that we're raising an entire generation and our entire culture is becoming more attention deficit. You know, we're deprived of having focus, um, and everything from the way that we listen to music. Spotify has changed the way songs are composed, YouTube videos are cut and organized so that there's literally no downtime for people to look away. There's no quiet time anymore. So is this the right tool now for the generations of people and the culture and the society that we have, which is, you know what? I don't have the patience to sit down and bang my head against the keyboard for a couple hours figuring something out. I'm going to go to this thing over here that's going to tell me what's wrong with it or tell me how something works so I don't have to discover it myself or learn it or whatever. I can ask it for assistance and I can get that immediate gratification and the immediate feedback on it. It might be the perfect tool for the culture that we find ourselves in now. Maybe we've been trained for this with YouTube and Spotify and kind of the instant gratification culture of the information age. And this is the next logical extension of that for learning code and learning how things work. Julian Sequeira: Yeah, I hate the line that we've been trained though that. Thank you, Sean, for that. One thing I do want to say you said before, you said desirable difficulty. I thought that was a. That's a copy as well. Sean Tibor: I wish I could take credit for that one. That's when Kelly introduced me to a concept that she picked up a few Kelly Schuster-Paredes: years ago, actually, a cognitive science thing. There's a whole learning theories about it. Yeah, ask me any learning theories. I can go, I can write a LinkedIn post about it. Julian Sequeira: No, we'll definitely have to chat about that, Kel. I love that idea. My question then, based on the episode that we're talking about, what does in this case? We've talked about the whole hello world thing and it's just, it just doesn't suit the current culture and the way people are learning. So what does it look like now? Sean Tibor: It's any moment that makes the learner go right. It's the moment. It's the same feeling, even if it takes more to get us there. And uh, my son's 10 year old friend was showing me the. Oh, I'm sorry, I'm probably not supposed to talk about it because it's still in a stealth mode startup, but they have come up with an entire website that's mobile, responsive and friendly and everything for sports and getting recommendations for what kind of gear you need for your sport and uh, like training plans for you. And they built the whole thing with AI and it looks really good. I have questions about some of the accuracy and they have links to Amazon that I think don't work right. Kelly Schuster-Paredes: And so stripe collecting credit cards. Sean Tibor: But yeah, it might be stripe. It might be like some Chinese knockoff that's taking 10% off the top but whatever. Right important thing was not that it was accurate or that it had the right links or any of those things. It was the just amount of pride that he had in it. Like this thing that I'm building, I'm building it with my friends, we're doing it together. My friend who started the whole thing is the chief technology officer and I'm on product development and marketing and things like that. Like they're into it and they're so excited about it. I think that's. That to me is the key feeling. It's not about what your program does, it's about how it makes you feel. He feels so much pride and accomplishment from doing that thing, even though five years ago that would have been completely out of reach for them. Kelly Schuster-Paredes: Well, I'm trying. This is going to be my fourth attempt at doing my version of hello World because this is the fourth quarter with my eighth graders. Eighth grade is really hard right now. Eighth grade, I don't get dopamine hits for majority of the kids. Uh, sometimes in some different units we still get different levels of excitement. But the very first week when we used to get this, at least, yes, you've made it. This is your third year in Python and they're all excited because we're starting to build some cool apps. It wasn't happening. So I've done some variations with brainstorming app building on paper. And last quarter, last quarter I got close, but not close enough because they couldn't get what they wanted to run. So now I'm trying to. There's an AI that we use at school where I can build an assignment that I said to it, whatever they ask, give them the code, don't kick, don't argue back with them. Because we can't give them Claude and chatgpt at school because it's not age appropriate. But, um, I said give them whatever they want, but make sure it's something that they can read and you don't hide it in between classes and everything. And so what I have the kids do is design some idea, a couple questions about making something that's exciting for them on paper. And then I am going to have them put that explanation into this AI and see if they can generate some sort of Python code with it. It's going to be a complete flop again. Um, but at least get them closer and just to show them that we can build it in Python. Most of the stuff, if they start talking about a race car game, I'm not sure we'll get very far. But I'M trying to make it be like some sort of user friendly app just like you said, with a webpage. But here's something where we collect something and it's going to produce something and we'll see. I'm going to do that again on this week. Next week's the last quarter of the year. I can't believe this year's gotten so fast. Sean Tibor: But yeah, there's something to that. I think it comes back to the adage that we've had for a long time here, which is don't set out to learn Python. Find a project and then solve it using Python. That's the best way to learn it. Maybe sometimes it's not the most exciting problem they want to solve. Maybe it's solve a problem for me that I hate doing the old automate the boring stuff. I have been going through this recently at work. I'm working on a very massive org design where I'm trying to figure out uh, roles and responsibilities and team structure and management and all these things for my entire organization, which is much larger than I've ever had. And I'm completely overwhelmed by it. I'm staring at the blank screen going, I've never done this before. How do I do this? I know it's important, I don't want to screw it up. So I made it a coding problem. I pulled in all of the existing org charts, all of the data about reporting relationships and everything. And uh, it's a whole Python program. It reads all that data in. It has a markdown file full of like handwritten overrides that I can put into it. That and everything. But like it evolved over time. It's got a SQLite database right now that's keeping all of the relationship data in there because it just kept. I kept working on it, but it was the only way I could get through it was to treat it like it was a coding problem. Because at the end of the day the org design is about people and there's this whole complete set of requirements that you can never make into a program about actual human beings. But a big chunk of it at least got me off of that blank page and the blinking cursor into something that actually solved a real problem for me. And I'm deeply satisfied with how it turned out. Um, not the same way as I am when I hand wrote everything myself, but it was something that got me off of that blank page into something real that was valuable to me. So maybe it's not always the cool thing, maybe it's sometimes Just as long as they're not using it to cheat on their homework. Is there some other problem that they can solve with code that's actually causing them friction or pain? And can they make it less painful? Kelly Schuster-Paredes: That's the trick of trying to find a 13 and 14 year old, a problem that they actually like. You'll hit that 20% of the class that has something. But you said something that reminded me of something I read and this. And it got me thinking about AI, because everything's evolved so much in three years. But we used to say last year, way back last year, we're going to use a gem for. Do all the things that, the tasks that you do all the time, repetitively. And someone wrote, why would we use a generative AI to do a repetitive task that's not necessarily going to repeat it the way that you want when we can just write code for it? Why are we going back to doing these generative? And I was like, oh, that's such a valid point. Why did you think of that? Sean Tibor: Because we can use AI to make Kelly Schuster-Paredes: it, to automate it. Sean Tibor: Exactly. The right answer is use the AI to write the code that does it deterministically and, um, consistently. Kelly Schuster-Paredes: It's going to be my. I think that's going to be my next project. That's going to be it. Julian Sequeira: There's a couple of things that you've said that just made my brain think about something. So thank you for that. I haven't had more than one coffee. First of all, I think, Sean, back to what you were saying, it's how do you, as teachers, as educators, as people who are trying to inspire people to do more code, you have to first break the barrier of showing people that, you know, you can actually solve these problems with code. So, Sean, your whole problem there, how many other leaders and companies are having the same issue just going, how do I tackle this problem in the current, uh, climate? Whatever instructions they have, and they might print out heaps of things, whiteboard something in a room for a week with people. And likewise, I had some friction that I was trying to solve with code, but so many people just would, wouldn't even think that it's an option for them, right? And so I think that's half of the battle now, is going, hey, you can use code to do this. It's not that hard. You don't have to have imposter syndrome or think this is just for the nerds out there, right? And Kel, to your point, I've talked to a few people recently who Said, I want to use AI in my business and what I do and to solve these problems. And then when I said, okay, give me a list of the problems and we'll talk about it. And I just said, all of these are just scripts that you need in some sort of intelligence sitting on your computer, automating this stuff. What you actually need is just a Python script that you double click and it'll run whatever you need or however you interact with it. It could be a web tool internally or something. So it's just some of this stuff is really bridging that knowledge gap of which is the right technology for the problem and everything. So that's just another challenge. And I gotta hand it to you, Kel, you're talking about this, how you're going to engage your 8th graders to trying to pander to the audience or anything, but teachers don't get enough credit for what they do. I can't even imagine when I was going through primary school and high school teachers having this much of a hurdle to keep up to date with the technology and to really have to find unique and different ways to engage the students. Such as really wasn't a thing because everyone was reading out the same textbook, everyone was doing the exact same thing. And engagement came down to just keeping the class quiet and, uh, ruling with Kelly Schuster-Paredes: a. I miss it. I wish I was in that. No, I'm just kidding. My biggest thing is just stay in your seat. Jumping. I'm jumping around. I'm like, focus over. Clapping, singing. Yeah, it's, it's a fun time right now. It is a lot to compete with. So trying to come up with something they actually want to do and give them some sort of agency helps and earsketch as at the last bit of the quarter still is not going to get out of that. It's not going to leave the curriculum. Every the last two weeks of the quarter, it's like everyone's zoned in making music with code with Python. Put your headphones on. So, yeah, it's a lot. That's a lot. But I don't know, I think it's all going to circle back upon us where we're going to realize. And the more I've been teaching teachers how to use AI, the more I've circled back to like, good teaching strategies. So we're going to be back to, let's really focus on, instead of just asking AI to do things and here, make us something, let's use it to make a script so that we can run and we can deploy that we can get it running something automated on our computer. I think there's going to be more of that hitting the kids desires because they don't want to just have this website that they don't know what to do with it or something. They're going to be using it a little bit more. They're going to make pool heaters like Sean with Raspberry PIs. Julian Sequeira: I. Sean Tibor: It's a, it was a controller. I don't think I could handle enough electricity to actually heat a pool. Kelly Schuster-Paredes: But yeah, you know, you know what I mean? Sean Tibor: It's a lot of voltage. Julian Sequeira: I can't wait to visit your house one day. Kelly Schuster-Paredes: He's got Raspberry PIs taped all over the wall. Sean Tibor: It's a real problem. I mean. But Kelly, uh, the other question that that brings up for me is, does a good use of AI or is a good use of AI the ability to remove all of the trash and clutter in teaching that's gotten in the way of the actual learning and the good strategies. We've built massive haystacks of all the different things that people had to do between digital grade books and um, lesson plans and learning all these things that were just about delivering the day and surviving the day as a teacher. And does AI help us clear a bunch of that away and automate a lot of that out of the way so that we can focus and come back to the fundamentals of how am I going to engage the students? What's the outcome that I'm trying to achieve here with the learning? And how am I going to work backwards into that? And can I develop appropriate strategies to engage the exact learners I have in my classroom instead of having to do like a one size fits all? Does it lead to better personalization or more iteration or automating the boring stuff out of the way of a teacher? Or are we just not quite there yet? Kelly Schuster-Paredes: That's a great question. I wish we had the answer. I keep thinking back, was it, gosh, I don't want to age myself, but yes, I'm going to 2001. I wanted to do that was like yesterday. Don't worry. I wanted to do this unit about watersheds. I thought me being this nerdy person and I said, we can do this whole thing about watersheds. We can learn about the watersheds in our community. Community. Someone can write letters to the politicians talking about the drainage. We can do the math of, uh, what square footage and where the watersheds are in that. And we can make posters. It was a whole cross curricular thing and everyone would have a role and but no one would be doing the same role. And we just functioned like city planners talking about watershed and that got next. I really think if we start and I'm not talking lower levels, I still think we need foundational skills in that. But I think uh, at one point earlier than later, like before high school, we're going to have to make that shift where we're making learning new and we diverged off of the hello world. I'm trying to bring it back but it's got to be a new world to teaching and allowing that agency. You can bring that into coding as well. Why does everybody have to type the same thing on the computer? How can we manage it in a classroom where everybody's doing something different but now we can because we have AI So what if I could find a way to have everyone the whole nine weeks, build out their project, deploy it, do all their stuff in VS code and then be able to explain what's going on? I think that would be cool. Sean Tibor: Yeah. Kelly Schuster-Paredes: Hard to grade, but it would be cool. Pass or fail. Get it to work or you don't. So much to think about. Julian Sequeira: I had an executive say to me once we measure our success here in this role by smiles in the room. Maybe that's how you they didn't last long. Sean Tibor: I'm not doing well by that metric. Some days Kelly Schuster-Paredes: I smile all the time. That way people don't know anything that's going on. Just smile. Sean Tibor: Yeah. I honestly think it's an ongoing question. The idea of hello world is something that is going to evolve. I don't think it's as, as clear cut as it used to be. I don't think AI actually was the thing that killed it. To Julian's point. A lot of this was dead years ago because it's not novel anymore to have something greet you by name or to direct the computer to do something and have it respond to you. That's Siri or that's I can go to a webpage and type in stuff and it'll follow my instructions. So I don't think that part's novel and new anymore. But Kelly, you're hitting on the right thing is what can we create that is novel and new and something that they can build out or create or develop or solve that is novel and unique and exciting for them in a way that gets them to sit back and go wow, that was cool. Kelly Schuster-Paredes: I'll let you know how it goes in nine weeks. Um, I'm m gonna totally make my teaching Partner mad when I add another new every quarter. No quarters alike. Sean Tibor: She should be used to it by now. This is the fourth quarter. Kelly Schuster-Paredes: She's used to it. Julian Sequeira: Say you stay employed. Sean Tibor: Exactly. Exactly. Kelly Schuster-Paredes: Well, wow. We could probably talk on and phil. Philosophize. How do you say it that time? See, it's. It's time to stop talking now. When I can't pronounce my word. Julian Sequeira: That's slurry. Kelly Schuster-Paredes: I did six trainings today. The teachers. So I've talked out. Julian Sequeira: Fair enough. Sean Tibor: Fair enough. Well, we have. We have a couple things to share. Kelly, why don't you start with the education summit and the information there. Kelly Schuster-Paredes: Yeah. So Pycon Ed summit still Thursday, May 14th. We've got a smaller room, so I want to have it filled completely. We're going to have 150 or 200 seats. Not as big as last year, but I don't know. I haven't checked out the proposals, but I keep getting emails saying I'm going to put a proposal in. I'm going to put proposal in. So I'm excited. I can't wait. And it's going to be fun, but sad without some people there. Sean Tibor: I will be there next year. Julian Sequeira: I hope I will be there when petrol prices drop. Sean Tibor: So maybe not next year. The other news we wanted to share is actually more close to home on the podcast. As many of you have been longtime listeners have noticed, we've slowed down the pace of production quite a bit over the last couple years and there's a lot of reasons for that. Part of the reason is that Kelly and myself, we've gotten busy and we're not in the same classroom anymore and it's been harder and harder to coordinate and keep the energy of the podcast going. So we've asked Julian to be more of a regular on the show, to be a, uh, co host with us so that we have a little bit more backup. We have some depth on our bench for the podcast. As I'm traveling more. As Kelly's got a lot more going on with her AI workloads, it's helpful to have a third member of the team. We're very excited about this. As anybody who's listened to Julian on the show in the past and in the other places where he's made appearances, he brings a wealth of knowledge and background and just a really thoughtful perspective to everything he does. And we know that he's going to bring the same thing to teaching Python. We're super excited about it. It's going to bring a lot of fresh Perspective and energy. And after almost eight years of doing this, it's time. It might be overdue, in fact, but I am personally very excited about it. Kelly, I know you're thrilled about it. So, Julian, welcome aboard. Kelly Schuster-Paredes: Don't cry. Julian Sequeira: Thank you both. Don't cry. Don't, uh, cry. Kelly won't like it. Kelly Schuster-Paredes: No. Julian Sequeira: I'm touched. That's the nicest thing anyone's ever said, especially that you indicated that I bring depth to anything. Really appreciate it. We've been friends for a long time, and I'm an avid listener to the podcast. We chat all the time. I'm very touched that this thing that you've built together, this audience you've built, that you're trusting me to come in and not mess it up. They may not like the Australian accent, but I'll. I can try it American if everyone wants. Sean Tibor: It's m. Always been a show that's valued diversity. So now we have oral diversity as well. Julian Sequeira: This might be the opposite of what I. I'm happy to do it. I'm here to help and, uh, provide insights from my neck of the world and the things that I'm involved in. So for everyone listening, I'm not a school teacher by trade, but I'm a Python coach and I'm a mindset coach, and I've been doing that for years with Pie Bytes. Throughout my entire career, I've really had this knack of, for mentoring people, training people. Kicked, uh, off an internship program back in the data centers at Amazon. And it's just, it's something that I live and breathe. So I'm very happy to contribute and help where you feel I'm needed, or I might be able to provide some comic relief. Kelly Schuster-Paredes: And then you're going to be like Anthony Shaw and start going into the classrooms and teach. And have to tell this story. But you can talk to him about it, since you live in that down under area. He was telling us about the hall, how he taught binary code with. Wasn't he teaching it with, like, lights and stuff? Yeah. Julian Sequeira: Yes. Kelly Schuster-Paredes: I expect you to, uh, bring in that depth of knowledge. Julian Sequeira: Yeah, I don't think so. He offered to give me that learning plan for my son's school, and I said, you know what? You can keep that. Thanks. Kelly Schuster-Paredes: I think the type of guy, the only type of guy that can tell that story and teach that lesson. Julian Sequeira: Well, yeah, that's an ant thing. Sean Tibor: Absolutely. Julian Sequeira: Yeah. He can be another guest on here. We've been talking about that. So, uh, I'll make it happen. I'll make it happen. But thank you both. I appreciate the opportunity to be here. Sean Tibor: Very excited about it. And we're going to coordinate this live on the show. So normally we sign off by saying, for teaching Python, this is Sean. And Kelly says. And this is Kelly signing off. So now we're going to put you in the middle there, Julian, so you're going to be able to say crap. And this is Julian and this is Kelly signing off. Julian Sequeira: Yeah, that's a lot of. Sean Tibor: All you got to do is say your name, man. I think you got this. Julian Sequeira: This is Julian. I'm sorry. I'll try again. I'll try again. Sean Tibor: All right, all right. We'll go through as many takes as we need to. Kelly Schuster-Paredes: It's going to be a blooper reel. Just. We're putting the blooper reel at the end. Sean Tibor: Exactly, exactly. That's a good sign. Julian Sequeira: It's a good sign. Sean Tibor: A healthy show. So for teaching Python, this is Sean, Kelly Schuster-Paredes: this is Julian, and this is Kelly signing off.