Transcript#

This transcript was generated automatically and may contain errors.

Welcome back to the Data Science Lab everybody! This is a place where we get together every Tuesday to pair code as a community and we have a different Data Lab Manager every week. Today I am joined by my co-host Isabella Velazquez. Isabella, would you like to say hello?

Hi everyone! Thanks for joining. I'll be posting the links.

Yes, Isabella is absolutely on top of links in the chat. So if you are wondering what we are using or talking about today and you think that there might be a link, give it a second. Isabella will probably put it in the chat. She's so fast and amazing. And I would love to introduce our Lab Manager for today, Joey Marshall. Joey, would you introduce yourself?

Yes, thank you so much Libby. It's nice to see everyone. I'm Joey Marshall and this is Tofu. You can't tell that he's a cat because of the shape that he's in, but he's typically a cat.

And collectively we are VP of Data Science at Verisight, a survey and public opinion research firm. I know a lot of you though through the Posit Data Science Hangout community, just because I lurk in the Discord. And so for the longest time I was a Data Scientist at the Census Bureau. So Joey from Census is the same as current Joey. I don't know, we change a lot over the course of our lives. Am I exactly the same? Kind of.

We're a million different people. Well, I bet a lot of people will remember you because you ask a lot of great, great questions at the Data Science Hangout. And I have been told before, I have been sent many a DM. Who is that guy who just asked that question? He has a podcast voice.

I'm like, he sure does. Joey and I need a podcast. Well, I will say Tofu is adorable and makes me want to ask Lauren to share her cat Nightmare in the chat somewhere. Share us a picture of Nightmare, Lauren, because Lauren has a very, very adorable black cat too.

Introducing today's topic: Claude Code

All right. Well, Joey, today our plan is for you to walk us through how you use Claude Code. We do not always talk about AI and LLMs and stuff on the Data Science Lab. We cover all kinds of topics. Today we happen to be talking about AI. And Joey, you have been using LLMs regularly for your data and other workflows for how long?

Oh, a long time. I mean, definitely I was using the early LLMs before there were transformer models to do work, before they were any good, really. But that was kind of in the before times. Ever since, you know, ChatGPT and Claude and Gemini have gotten good, I've increasingly used LLMs both to build things for people, either employers I've worked for full time, or I've done in the past some side contract work for employers who wanted things built with LLMs. So that would be like fine tune an LLM or do a wrapper around an LLM in order to create something for our customers. So that's one use case. But then the other use case, which is kind of more central today, which is pair programming with LLMs, like using Claude Code to help me build stuff, which I increasingly do. And that's probably more relevant to today's chat.

Yes, absolutely. And I say we go ahead and start screen sharing and start working through what we're going to do today, which is working through a tidy Tuesday data set. But giving Claude a lot more free reign than I would in an analysis, this is going to be fun. And while we are doing that, Noor has our first question today in the Discord, which is for someone who's bearish on AI, what benefit does using something like Claude Code bring? And second question, which I want you to answer first, what's your favorite pie?

Oh, hey, Noor. My favorite pie is that pie that you posted in the Discord last night, because it's the most beautiful pie I've ever seen. And I really like your question on being bearish about AI. I'm even reluctant to assume that identity myself. And that's something I'll try to be sensitive to as we go through the... Like today, we're going to intentionally use Claude Code and really push it to the max and try really hard not to do anything hands-on. That's not a representation of the way that I always work. But I think some reasons to be skeptical, especially for us data scientists, would be statistical errors, whatever you build needs to be accurate. There are security risks like giving secrets and your API keys and tokens to Claude Code and just exposing your file system to Claude Code. It's a cli tool. Those are all reasons not to be bearish about AI. But there are a lot of good reasons actually to appreciate working with it too and to try and strike a balance.

And for me, that's that... I mean, obviously, I can work a lot faster with Claude Code. That's not always a virtue. Sometimes being slow is preferable. But if it's something that needs to just work, let's say it's like a front-end tool. I'm not a front-end developer. I don't need a production-grade front-end tool. I just need a clicky interactive thing that will work as a proof of concept. And then I can hand it over to real software engineers to build that kind of thing. Then AI is great for that. First of all, I know it's not making any statistical errors because I'm not asking it to do any statistics in that case. I just want it to mock up like a simple front-end, which it's great at.

And you can just fly through tasks like that with Claude Code. In more sensitive stuff, you have to be a lot more careful, build in a lot of unit testing, look at a lot of stuff hands-on.

But the other thing is always learn a ton from Claude Code. And I think this is true of all the AI agents. They're great teachers. Like what Wikipedia used to be. I still love Wikipedia. But it's a great jumping-off point. Don't get me wrong. I still use Wikipedia all the time.

No, I mean, me too. Right? Yeah. I think that we will demonstrate some of this stuff as we go through today. And Kieran just said in the chat that Claude's loop where it can be permitted to use local resources to write, to run, to test code on your machine, to read output from it, turns out to be really a powerful way to iterate on things. And I think we'll do a little bit of iteration today and demonstrate that.

But the other thing is always learn a ton from Claude Code. And I think this is true of all the AI agents. They're great teachers. Like what Wikipedia used to be. I still love Wikipedia.

Setting up the terminal and tools

So the first thing that we will do is I'm guessing tell Claude what we want to do. And what are we looking at here? We have your terminal is up and we don't have a Claude open yet.

That's right. This is just an empty terminal. And for, I'm not using the, okay, I'm on a MacBook Pro so you know what computing system I'm using. And my terminal here is through cmux. I have a list of stuff that I use and Isabella and Libby can share that at the end. I've shared it with them. I like cmux because it has nice vertical tabs. Check this out. I can spawn up all these terminals and have them all doing different things in these pretty little vertical tabs.

cmux also hooks into Claude's hooks. So you can get like desktop notifications whenever your Claude's done doing stuff, whatever. I don't sell this product. It's free. cmux. Okay. We're looking at a bare terminal and I have some terminal aliases set up, which are, if you don't use the terminal a lot, you'll have some configuration file, depending on your operating system that can, you can enter some characters and they'll translate to other kinds of characters. So for me, the terminal alias dv just puts me in my dev directory. So I'm on the directory like joey slash dev.

Are the letters on my terminal big enough to see? That's a great question. Could you do one additional zoom in, which might be like a command plus sign as well. It looks like command equals. Oh, command equals. Oh, plus is the same thing as equals. All right. Promise I'm a computer man. This is good. We had some feedback on the YouTube that was like, why is the screen so small? So we really want to make sure that people can see this later.

Also, if you want to watch somebody bumble around and not always know what they're doing, I'm. That's the point. That's why we're here. Okay. So remind me what the alias dv does it one more time. It just CDs into, you know, whatever my directory is, you know, users, joey dev, I'm about to use another terminal alias, which is C C D P. Okay. So normally you would start Claude like this. You got to spell it right. That was collude. So you would start Claude like this, and you can have some flags, like if you want it to be able to access Chrome. And if you want to dangerously skip permissions, it's sometimes called vibe coding mode, you know, whatever. So I have a terminal alias called C C D P that is going to boot up Claude Code for me.

Managing secrets with 1Password

And you'll notice it just asked me for my fingerprint password. That's because I'm running Claude Code inside of a 1Password command line tool. This is also in my list of resources. 1Password, you know, it's just a password manager, but it has a secrets manager that can interact with your command line. You can run Claude Code inside of it, and that lets you do really useful stuff, like keep your passwords secret. So all Claude Code sees is a link to a password, not an actual password. I'll say more about that later. Now we're in Claude Code, and we're going to let me just grab this data and pull it over. I've never worked with this before. It's about frogs. That is about the extent of what I know. So we are going to do this in real time.

Can you see my whole screen? I'm sharing like a reading from a GitHub right now. Yep. We're looking at the Australian frogs data set from tidy Tuesday, which is I believe week 35 from 2025. I think it's last year. That was an incredible display of memory. Can you tell that I work with tidy Tuesday a lot?

I just casually dropped an extremely exact date. Okay. So we've got this frog ID data. This gives us instructions on how to read it using R, and we're kind of going fast and having fun. So I'm just going to copy this URL, and then we're just going to drop this into Claude Code and ask it to explain it. I'm using voice dictation to communicate with Claude Code, an app called Wispr Flow. It's also in the list of resources. I'm doing a live demo of Claude Code, and we need to build some kind of data product within about 45 minutes. So don't let me down. That's the first thing I'm going to say to Claude. I really need it to be on its best behavior. The first thing I want you to do is to explore this data set. Tell me about it. Make sure you can read it and just explain the data a little bit to me, and then we're going to drop this URL in and let that cook.

Okay. While this is cooking, I want to quickly explain what I was saying before about wrapping my Claude Code command in the 1Password password manager. Okay. So often when you, okay, it's doing stuff. Downloading CSVs. Great. Often when you work with an AI agent, if you want the agent to be able to do things autonomously, you need to be able to share secrets, API tokens, credentials for stuff, URLs to things that may not be public, whatever. And it's common, I think, in practice to drop those in like a .env file, .env in your directory. And Python has libraries. But the problem is, you've exposed a list of secrets in plain text on your machine. If it's a shared computing environment, that's really risky. And also, if you accidentally commit that, that's a real problem. So I want to talk about how I manage secrets. But I'm going to take a breath because Claude's done and we're going to look at Claude.

Exploring the frog ID dataset

So put an asterisk in what I was saying about secrets and let's see what Frog says. Frog. Let's see what Claude says. This comes from Frog ID, a citizen science project where people across Australia record Frog calls via a mobile app.

There are two files. Frog ID data.csv. It's got about 140,000 records. Each row is a Frog call recording. With lat, long coordinates, we may can do something geospatial. State. And which I guess are top level geographies in Australia. And recorder ID. Okay, cool. 186 unique species. That seems like a lot to me. How many? I mean, it's a lot, but it's a diverse place. By the way, as a little bit of background, GIS stuff, spatial stuff, is what Joey does. It's like Joey's specialty. So this would be a fun thing to do if Claude could help us. Oh, well, now I've been overhyped. Please manage your expectations. I'm just kidding.

New South Wales dominates with 43% of records followed by Victoria and Queensland. Okay. And then we got Frog names, which just maps the scientific names to common names. Okay, cool. What kind of data product are you thinking? An interactive map dashboard analysis or something else? So here's what I'm going to do. Let's do some interactivity. I'm going to get Claude to give us a few ideas and then we can all decide as a group what we want to do. It's a safe place. Right? We're among friends. We are among friends.

And while Claude is cooking, we will talk a little bit more about secrets and then we'll get to David's question that happened in the chat. But I'll let you talk because I know that you're going to dictate. I'm too lazy to press buttons on my keyboard. Actually, if you have Carpal Tunnel, which I do a little bit, the voice dictation can really help. So I voice dictate almost everything to keep my wrist from being

messed up at work. So it's like an ergonomics thing, too. I want to let the audience decide what project we will build exactly. So why don't you propose three interesting projects? It could be stuff like overlapping the frog data with like population demographic data from Australia. That's just one example. Maybe something geospatial and interesting. Basically just stuff that you think is feasible that you, Claude Code, can help me build in just a few minutes. And that is interesting to look at. So a web app is a good idea. I have access to Vercel if we want to build a web app and deploy it in real time. Maybe we'll create a GitHub repo, deploy whatever we build to Vercel, and then I can share all that stuff with the audience. But give me three of your best ideas and make sure that they are interesting and feasible.

Just stream of consciousness garbage words to Claude. We all thought you were talking to us. We all instantaneously forgot that you were talking to the computer and we were like racking our brains to figure out suggestions to give you. People in the chat were like, what are we going to ask you? Oh my God. So that was amazing how instantly our memories just blipped and we were like, no idea what he's doing. What is he talking to us? No, he was dictating. Everybody was dictating to the computer. That was amazing. And actually the reading through the transcription of what it gave you was amazing because mine would have been filled with large silence gaps and ums and like all kinds of filler words. You're doing great dictation.

I'm actually not. A lot of that is in whisper flow. It's again, like everything is AI. It's like AI all the way down. It's like an AI voice dictation app. So it cleans up the dumb stuff that I say.

We are all just cracking up in the chat because we were so sure you were talking to us. Okay. So David's question that he had asked, I think we're going to leave it with the solution that we have in the chat, which is actually one of Simon Couch's blog posts on LLM evaluation. Because David had asked, have you done any analysis of how the different generative AIs perform doing the kind of work you're doing? I think that the analysis part, Simon Couch, just like goes way above and beyond anything that any of the rest of us have done. But what is your general vibe sensing on the different types of models specifically like Codex versus Claude for what you'd like to do?

That's a great question. I should have said at the beginning, I'm using Opus. I think 4.6 is the latest Claude code generation here. Codex is getting better. I tend to use it less for vibe reasons because it feels less, let me give a more intelligent answer.

Definitely read Simon's analysis. I do most of my work in Claude code using Opus. You burn through tokens really fast that way. Sonnet is fine for light engineering stuff. I tend to only use Haiku, Claude's smallest model for LLM tasks, like when I want an LLM to do something, like read some words and extract something or whatever.

I did build an integration once, and you can find it on my LinkedIn and GitHub and stuff, where I could ask Claude code. I could give Claude code the autonomous ability to get a code review from Codex on the same machine. I liked that for a while until Claude code's own code reviewer sub-agent was created, I think in the last generation or two. You can ask Claude code to deploy its own code reviewer. It's a code-reviewer sub-agent, which is a much better code reviewer in my opinion. I just haven't used Codex all that much recently. Sorry that I can't give you a better answer. I'm using Opus right now.

Prompting strategies and voice dictation

That is totally okay. And Becca had also asked a follow-on onto this whole prompt, which we are about to look at and go over the results of. She asked, with your Claude prompts, what components are you intentionally saying versus what did you feel free to be more liberal about? And I'm guessing that means, like, let Claude interpret on its own instead of specifying. And are there any aspects of your prompts that are directly influenced by your training as a data scientist? Oh, that's a great question. I used to try to be very precise in my prompts. And just anecdotally, I've learned that the more words, the more human language I can give agents, the better they perform. I mean, think about what it's doing. It's taking your prompt, it's getting an embedding, and then it's moving all that context back and forth and back and forth every time. And so, like, in terms of it providing generating relevant text after what I've said, the more I can give it, the better. As a matter of fact, this is just, like, a general, I think, best practice, is I tend to have Claude code write as much documentation, maybe more, as it does code. In a second, we'll use planning mode, where it sends up a whole bunch of agents, does a whole bunch of reading and writing. Before it does anything, I'll have it write a bunch of documentation that we can reference. The same principle applies to prompts that I give it. I tend to just say a lot. And if you're unsure, just say even more. And that's why voice dictation I really appreciate. So I'm less precise and much more verbose with my prompts than I used to be.

I tend to have Claude code write as much documentation, maybe more, as it does code. I tend to just say a lot. And if you're unsure, just say even more. And that's why voice dictation I really appreciate. So I'm less precise and much more verbose with my prompts than I used to be.

I think we'll move on in just two seconds after we stick in Zach's question here, which is how does the computer know that you're talking to the AI versus just talking? Like, is there a button you're pressing or a hotkey? What a good question, Zach. That's I bet there's an accessibility feature that would let you, like, look at a different screen. Right. Or be like, hey, Claude. Yeah. Yeah. It's like the yeah, you can set a hotkey. So I've assigned a hotkey that I hold down. You can choose a hotkey. Thanks, Zach.

Choosing a project and reviewing Claude's proposals

Well, Ijlal, I see that you had a question as well. And we will get back to that in just a second. Let's go ahead and review this code. Because we already have people voting in the chat about what option they like. So many people have said they like two out loud. But if I look at the vote Isabella put up with emojis, oh, wow, the votes just changed. We have nine votes for one, seven votes for two, and two votes for three. So let's review. People like frogs after dark. Oh, now they're tied. One and two are now tied.

Let's see what we've got. It gave us three proposals. All feasible as a single page web app that we can ship to Vercel in under 30 minutes. What the heck is Vercel, Joey? Oh, it's just like another back end as a service app. I have so many of these goofy things. But Vercel, okay, there are like a million services that Vercel is probably one of them. I actually don't know what Vercel's back end is. And sorry. Like DigitalOcean? That's another example. Okay. For databasing, like I use Supabase a lot, like Railway for workers or GitHub Actions. All these back end as a service things, a lot of times what they're doing is reselling you AWS. But what you're paying for is a ton of convenience. So they're really doing some work to design away a lot of the friction in getting a working thing online. So Vercel is just another hosting and deployment environment. But it's super tailor made for Python, which I work in a lot. And so you can just like put together and deploy a Python app and then layer on a little front end and it kind of just works. So I have a Vercel token anonymized in my environment here. And it's just like a quick way to get up and running. For your own org's enterprise solution, your org probably has something better with like software engineers on staff who can help you do it. But if you're just like doing stuff as a person and for proof of concept, all these services are I think pretty great and they're all pretty cheap. Usually everything is like 20 bucks. But then it's like streaming services. You have $200 a month of $20 a month stuff, whatever.

So who's the winner here of the options? Isabella, who is our winner? I think as it currently stands, it's option two. And apparently we get a pie. Yeah, well, no, I was hoping for I mean, like, nightlife heat map like sounds really cool. But I like that Claude is proposing like a radius query. I think that means we can make a little circle and then learn stuff about it. That's how I'm interpreting that. So let's do it. Okay, let's do option two.

Planning mode and deploying to GitHub and Vercel

Here's what we're going to do. I'm going to do this. Okay. Option two is the winner. Would you please go into planning mode and do deep research on what we need to do to build and implement option two. At the end, I want this to be a GitHub repo that you can deploy to my personal GitHub. By the way, look in your global Claude.md file for information on how we manage secrets. You should be able to see my personal GitHub pat in one password. And you should be able to use that to deploy to GitHub. Also, I would like for this to be an interactive web app. My Vercel token is also in one password. You should be able to get it there. Otherwise, I want to make sure that whatever we can build is feasible. You've thought through both the data that we need and the tech stack. And also the aesthetics. I would like for this to be well, it's about frogs. So I would like for this to have a cottagecore aesthetic. I'm going to ask the audience for a color scheme. And I'll provide you a color scheme later. Go into planning mode. Think this through. And come back to me with an implementation plan.

Oh, my goodness. You guys, we need color schemes, stat. Raha says cottagecore versus just green. That's a good question. She says, I think Joey would be a better interior designer than I would. Okay. We have so many questions. So let's get to them as fast as we can. Because we only have 30 minutes left. And I think that we can probably hit a few of them pretty easily.

So, let me see my little questions here. We had one that was from Bruno that said, can you please ask Joey, do these voice to text things require a lot of hardware to work fine? Whisper flow is cloud based, I think. It's a good question. I don't know. I don't know how well whisper flow is optimized and whether it uses any kind of GPU acceleration. But you know what? I bet it does. Let's, I don't know. Let's Google it. I can Google it. Can someone Google that? Does whisper flow use GPU acceleration? Someone in the chat can do this.

Okay. I'm using, like, an M4 MacBook Pro. 48 gigabytes of shared memory and, like, it has, like, a 12 core CPU. And the GPU is, I don't know, pretty beefy, you know? So it's like, but you know what? I also run whisper flow on my piece of trash Windows machine and it works fine. So it must be pretty well optimized.

I love it. We, Emily has put a cottagecore picture with literal hex code colors picked out of it, which is amazing. We're going to go with that. Lots of people are, like, raising their hand and saying, yes, that. I don't know if you have that up, but I can stick it into the chat of Zoom as well so that other people can see it. Oh, that did not format very well, Zoom. Oh, I see it. No, I see it. It'll help Joey see it, maybe.

And then I wanted to ask, also, Zach says, Joey really does have a podcast voice. I think he could voice a nature documentary. I agree. We're going to collab on this. So JPS had asked, and I believe that that's Jacob Patterson Stein. Good buddy. Do you have advice on setting up agents that are working on the same folder? I worry that if I have too much going on in the same folder, it will create some sort of recursive confusion. No, it can. Yeah, that's a real thing. There's a slash command, by the way, in Cloud Code, which is slash BTW. It's, like, just to let it know, while you're working, here's something. I'm going to say, here's a screenshot with the color scheme and aesthetic vibe that I want. And then, yeah, it's just control V to drop that image in there.

Okay. So, oh, this is probably not in my I can see the screenshot. But unfortunately, I'm not able to view image files directly in this context. I can only see text content. Oh, maybe that's because it's in planning mode. Oh, well, that's good. We have the text. Emily pulled out all those hex codes for us, which is fantastic. Okay. Right above the image, if you want to grab those. And Isabella also put them in the Zoom chat in a very nice formatted way.

Nice. Okay. Let's just drop those in. Or just describe it. Space enter or escape. I pressed space and probably shouldn't have. Now it's just kind of rolling. It's moseying. These color names are amazing. Noir mystique. Avocado stone. Desert devil. This is now ASMR.

I'm sorry. I won't do that again. Okay. So, Noor had also asked while we're cooking here, she's like, I know 1Password is a password manager, but how does Claude have access to it?

That's a great question. Okay. Let's do both. Whoever asked about conflicting subagents. JPS. This is important. You can ask Claude to spin up subagents or you can use the slash. Oh, look. Love that. Love that palette. Very cottagecore. I've updated the plan with your colors. Okay. Great. This is a real problem. So, you can use the slash agents command in Claude code to either define your own subagents or it has some prebaked subagents that you can use like the code reviewer and some other ones. Or you can just ask in natural language, please spin up agents to parallelize this task as much as possible. You know, whatever. And it will try to do that.

Those agents will Claude code is pretty smart about delegating its tasks to its subagents. However, if you have multiple Claude code sessions. I spun up a bunch of Claude code sessions here. And I had them all working. And they were all working on the same repository. They will absolutely conflict with each other. And if you're doing stuff like that, you should use git work trees. It's a little out of scope for today. That's like a whole data science lab using git work trees. But it basically provides like an OS level kind of barrier between the various agents. So, they won't conflict with each other. They kind of all get on their own branch and do their work. And then, you know, conflicts, merge conflicts are designed a way like that. So, git work trees are a thing in git. But Claude code is designed to integrate with git work trees. That's a great way to parallelize across multiple Claude code sessions.

Otherwise, for Claude just spinning up its own agents, like in planning mode, which we just did, Claude almost certainly spun up a whole bunch of subagents to do this. Just to do specific tasks. Oh, you go off and research this part of the tech stack and you go off and look at the data and you get off of whatever. And you can see they're little commands. You can see them as they work. But git work trees if you're if this is a concern.

How 1Password CLI injects secrets

Nur asked about secrets. Okay. So, 1Password has a command line interface or cli tool. And what that allows you to do is to wrap your Claude command in a 1Password command called oprun. And what that does is let me just show you. Okay. So, I'm in this I'm back on a terminal. We're off of Claude code. I'm back on a terminal. And I'm in my dev directory. I have all of my keys are in here. I'm not going to show them to you because actually they're not printed in plain text. I'm just going to show you the file. There's a file called env.env.1password. And these are my keys. But they are not raw keys. There's nothing this video can go on YouTube and no one can get anything from me here. They're just paths to my 1Password vaults. So, you set up a vault in 1Password.

paths to your 1Password vaults and then the op command that you wrap Claude code in and I just dump all that into a terminal alias more about this in the resources that will get sent around this will trigger something called password injection so Claude only ever sees this but when it runs a command the 1Password cli tool will replace this with the actual secret that'll get injected and then you don't have to work with bare secrets in a text file that makes sense

when it runs a command the 1Password cli tool will replace this with the actual secret that'll get injected and then you don't have to work with bare secrets in a text file that makes sense

amazing yes this is so helpful uh nor thank you so much for asking that question like asking what all of us would would like to ask but we're not um there are a couple of questions that I think can be answered in the chat if anybody wants to hop in there and help us out um like Aaron had asked is there an equivalent to the cli 1Password functionality and Apple passwords on Mac I do not know yeah you can use keychain it doesn't have the same I don't think keychain has the same password injection thing um but I do know that people use keychain with um AI agents so there's some way to do it I've never done it okay and then you're all asking amazing questions still but I want to hold them a little bit to the end so we can get through these because they're more overarching questions um that I think we might answer as we go through a little bit as well um also Amelia put a resource in there about Bitwarden um Bitwarden also has a cli

Reviewing the implementation plan

let's go ahead and review our implementation plan from Claude nice so after I've asked Claude Code to go into planning mode you can also there are hotkeys to turn planning mode on but you can just ask it in natural language and now it's put together a plan okay here are these colors that we asked for uh now it's going to tell ready to code oh I yes I feel ready um so context we're doing a live demo blah blah blah just moving through this here's our build order fine uh what we're going to do with the data uh okay cool um key technical decisions I don't we're moving fast so I don't care a cut list if running over time um and then okay Claude Code has written up a plan and is ready to execute would you like to proceed and then you'll be asked uh yes or if we have more time we might manually approve edits okay if I were working on this for real or um you know I might tell it to change some stuff change something about the tech stack I'm just gonna say yes and we're just gonna YOLO like yeet this into the AI sphere and just like see what happens

um so this is gonna be cooking for a bit and I'm happy to what were we gonna do something I think more people had questions that you were gonna have well Isabella has one Isabella you wanna ask yeah I saw that it gave the option to clear context and it said four percent what does that mean oh that's a great question

Understanding context windows

okay so let's get out of this actually let's open up a new terminal that's so nice that you're just like we'll just open up a fresh one it's so it's honestly so nice I'm gonna open up just like a raw Claude session and I'm gonna use this Positron too I'm like constantly just opening another instance another session another console that yeah Positron is great about that by the way I use Positron a lot is sometimes these days kind of my main IDE yeah me too I was in RStudio for R and Positron for Python for a good like year now I'm just in Positron I haven't opened up RStudio in a while but I still open it it's still my default for opening MD files so every time I click an MD file RStudio opens up and I'm like hey RStudio

okay so um I may be mansplaining a lot of stuff that people already know but no please explain it because it'll be great for I mean it'll be great for me selfishly I want to know all of these things I'm a very baby Claude user yeah and I got that message the other day and I was like I don't know what this means like will it delete everything that we've talked about or things like that no it's such a good question and it's a great thing to talk about while Claude is doing stuff

okay so what's an LLM doing it's not doing anything magic a lot of it you can think back to your early like text analysis and NLP principles an LL the a chatbot essentially is taking your prompt tokenizing it you probably remember from early days that a token is just a fundamental unit of language Claude GPT all the various LLMs tokenize in different ways they might be usually parts of words or even really Claude tokenizes really aggressively so even like a character space is like a token stop is a token like stop generating text those are all tokens but anyway it's going to tokenize your thing it's going to get its your prompt it's going to get its sentence embedding kind of its uh sort of semantic kind of like it's like it's like it's coordinates in highly dimensional semantic space that's not exactly accurate but it's a nice heuristic of how to think about sentence embeddings uh and it's working memory can only hold so many tokens at once and before last week Claude Opus 4.6 could hold a total of like 200,000 tokens so whenever you're having this conversation throughout these tokens are going back and forth between you and the and the LLM and its context windows filling up filling up filling up filling up uh larger than it can handle and when that happens uh it's a real problem um first of all anecdotally I see performance issues whenever my context window gets close to the end um but more more importantly when your context window fills completely up you can't prompt Claude anymore so it will automatically do something called compact the conversation which is it will write its own summary of the conversation thread and then clear everything else out of memory now you can trigger that compact yourself slash compact and you can even write your own summary of the conversation thread um now as of last week or maybe over the weekend uh the context window for Opus is now a million tokens it's 5x the size it expanded

saw a lot of chatter on reddit about that this weekend people are hyped I mean I'm hyped it's nice um and if you do slash context in a Claude code session you can see this really helpful I just love little visualizations in the terminal oh what creative way can we come up with to have little characters that mean stuff and make kind of a graph it's just like a fun data viz challenge uh so this is basically telling us that three units of this are consumed with its system prompt uh you know that's like stuff that Anthropic gives it don't be evil you are an AI you know whatever it has like a don't actively harm things yeah Anthropic has a philosopher on staff uh she researches like morality and well good yeah that seems I'm glad

um and then there are a few skills we haven't talked about skills yet but Claude code skills are um like markdown files that tell it how to do stuff that is useful to you yeah we can have a whole day science lab about skills probably oh time check we have 15 minutes left all right let's hurry up thanks hey frog finder is live at this url what already like it deployed it I mean I gave it my vercel token um okay messages this is stuff that I've sent back and forth to it you know whatever eventually this will get almost full and it has a little auto compact buffer so it doesn't let you go above this so when it's getting around here you want to compact your conversation okay uh github auto linking failed needs a vercel login connection for that github org oh that's right I haven't authorized vercel and github to communicate with each other but the deployment works via direct upload okay I can connect the repo in vercel settings later um okay well I mean let's look at the let's see let's see what it came up with

Checking the deployed app

people don't really say vibe coding anymore but if that word still existed this would be the most vibe coded anything has ever been okay let me grab a browser I'm so anxious about this

okay to get started oh am I like logged in or something this is actually the part where I usually fail with Claude is the deployment part where like I don't have things hooked up in the background for it with my pat and stuff like that um and things fall through the cracks but I'm curious about how this will work please feel free to log in on the side if you need to we're gonna give Claude a browser and let it look at the browser itself so I have a chrome browser with the Claude extension so here we go

when I visit that url I just see some stuff from vercel asking me to deploy the project I'm not sure that that worked can you tell me what could be wrong is the url wrong also you are authorized to use a chrome browser if you just want to go to that url directly and then you can tell me what the problem is

okay let's see what it's going to do okay it's going to do stuff in the browser we can watch it recently asked it to book some flights for me using the browser and it did better than you think

time check um it's left uh we weren't I think we finished the conversation about context right yes um and David had asked if that 1 million token change applied to all paid levels I don't know but I think yes I'm googling and that's why I said I think yes but not totally sure happy St. Patrick's Day everybody by the way I'm wearing green are you wearing green I'm very Irish

okay Claude is still thinking um and so I'm going to pop over to Joe's question in the chat which is so you can switch back and forth between the model during the same session depending on what you need

what you need or do you have to stick with a single model while you're in that one session

because you can slash model right Joey yes uh oh was the question can you switch models in the middle of a session oh um good question I tend not to do that either I don't know I don't know if that would wipe your context maybe does it does someone else know I actually don't know the answer to that if you know the answer put it in the chat um my gut tells me that you would have to re give it re give it all the stuff right you'd have to re-give it your context your cloud md file all that stuff I would think but I actively do not know David says it does for all paid API oh like you can't he's David is saying you can switch back and forth um that one was for the 1 million token expansion for the context window so thank you David for checking up on that

yeah Joe says I guess we could ask Claude itself we could I don't know this is something that we will have to try okay I think you know what's interesting is often the models don't know that much about themselves because they were trained on data in a world before that model existed Claude will try to reference its own docs but yeah I have a question from Isla curious to know what resources you would recommend to learn all of this stuff how did you learn all this stuff Joey oh logs YouTube videos trying and failing yourself well you could join the data science hangout discord yeah so the discord where you are right now is a great place to hang out with people we have an AI and LLM channel when I remember asking Hadley this question not that long ago and he was like blogs I read a million blogs and I think that that's a great answer and I've watched a million YouTube videos as well Isabella how about you what have you learned from yeah I also find like Anthropic has like a course marketplace and you know they're fairly short and um from like the foundational so I've taken a few of those as well oh my gosh okay

all right uh I'm gonna drop this URL in the chat so other people can go there too I guess if everyone goes there you might blow up my uh $20 a month hug of death everybody yeah exactly uh okay what can we see

Exploring the frog species map

oh there we go okay I can click on a circle oh it just makes an automatic circle I kind of wanted it to I kind of wanted to draw a circle oh I can increase the radius here we go there we go and then this tells me the species found oh and it says at the top it says 41 species found within 90 kilometers of where we are centered which is kind of right around Brisbane

huh look at all of those little those little froggy dudes

oh they're so cute so this kind of worked can I click on that does this do anything no it does tell you how many observations there are though for each one so for red tree frog we have 27 observations tells us its uh scientific name as well as it has its common name as the the bolded heading super cute and we do have our cottage core very cute desert temple vibe stuff going on

yeah we did this in like half an hour because we talked for a lot of this right and in the beginning and we're 52 minutes in we have eight minutes left and we have a thing that is mostly working as long as we can verify that it is correct right yeah I mean maybe not like ready to you know um but I mean yeah but for a POC proof of concept this is great I think so yeah I don't hate it yeah Amelia asked can the next iteration of this map use frog emojis instead of dots for the location oh uh okay clicking maybe in the next six minutes we can ask Claude to do that

it might take too long to cook oh to use frog emojis yeah instead of those dots oh oh I like that can you use frog emojis rather than dots and then redeploy oh also did you push this to my github I want to share the github repo with people this is how much I don't know about what's happening under the hood right now like I don't even know that this has been pushed to my I normally don't work in quite this sloppy of a fashion oh yes it is on github okay let's see

oh no it pushed to my um work github not my personal github no I'll put it on my personal github and then I'll put it in the discord yeah Anna says it's kind of weird that the dots are equally spaced like put them in a grid I wonder that's weird I haven't looked at the lat long data but I'm wondering if that's a limitation of the data or if it did that on its own and it just like uh aggregated within a grid I don't know

that's a question we would want to ask it you know like about the choices it made and we'd fine-tune this as we went you know for a real thing

um does anybody know if the images are in that data set for the frogs I don't believe that they are

oh like where did it get these images yeah do you want to ask it like hey where did you get the images for these frogs that's a good question

so you get the images for the frogs

oops we all talked at the same time so Joey asked it um where did you get the images for the frogs out loud with dictation and he used the slash command btw to insert something while it's cooking oh we got them from wikipedia a few folks mentioned that it was actually in the plan I um that's oh we just missed it because we were we're trying to go fast sorry yeah thanks I had thanks Aaron

Aaron let us know that it was it was it told us in the project specs it was going to pull them from wikipedia okay yeah Amelia same question I'm really curious about how accurately it did it though um that's the kind of thing that when I am working with Claude I am constantly cherry picking to check myself I'll go like grab five of them and then I'll go look them up myself um and I'll try to make sure that I'm verifying that things are coming through correctly so we would also build a whole bunch of unit tests to double check its work we are for clarity we're going so fast right now right because we had an hour more like 50 minutes to do this stuff we wanted to see how much we could get Claude to do in sort of a one shot here frog emojis are live

I'll go like grab five of them and then I'll go look them up myself um and I'll try to make sure that I'm verifying that things are coming through correctly so we would also build a whole bunch of unit tests to double check its work

are they still in a grid or if we zoom in are these randomly spaced froggies they are still in a group how curious but some of the frogs are larger and that's why my question was like maybe it's aggregating them into a grid and it's basing the emoji size on the n value of how many frogs there are I don't know Tom says I don't like the frog emoji it's like a space alien I love it plague of frogs says newer this is very biblical okay

oh Darren I'm so glad you're here oh they're just they're just put in um uh grid cells so a tenth of a degree oh we would have to get into the details of like uh what is that in this projection um yeah I don't even know how this map is projected it's a leaflet map so maybe it's Mercator by default um so yeah Mercator by default I like the little explanation though small medium and large small fewer species bigger moderate diversity um yeah this is something we would have to dig into which oh it's guessing 11 kilometer squares uh that's does that vary going away from the equator probably I don't think these would be equal area grids ah that's getting into the weeds of like geography stuff whatever it doesn't matter um plus Australia is a huge place right amazing well this was this got us so far we learned so much there are a million links in the chat but also when this video ends up on YouTube um if you are looking at the YouTube videos all of the links you could possibly want are in the description um they're usually just like stacked with with links and if anything is ever missing from those send me a message on discord and I will go fix them and just add stuff to the the descriptions in YouTube thank you so much for hanging out with us Isabella thank you for hanging out Joey I hope you had a wonderful time thank you so much for sharing your knowledge with us this was fun thanks everyone yeah everybody say thanks to Joey in the chat this was so much fun I learned so much bye everybody