General Assembly with Niklas Fischer

Speaker A: Hello and welcome to this week's episode of Would you Data Scientist. I'm Wendy Gannon and I'm joined this week by Nicholas Fisher. How are you doing, Nicholas?

Speaker B: I'm good, Wendy. How are you doing?

Speaker A: Yeah, I'm really well, thank you. This is the second time we've done this, isn't it? I, um, foolishly, we finished our chat last time and then I pressed the record button to stop recording and it went, you are now recording. So we had, we, as you said, we just had a nice chat. Uh, um, but bless you, you're back and you're, you're, uh, here to talk about, um, data science. So let's kick off firstly with what your favorite piece of tech is.

Speaker B: Yeah, um, so I think, yeah, for me it will just be the Internet in general. Um, I think for me, because I've moved around now quite a bit, it's just the best way of just staying in contact with, you know, all the friends you sort of met, um, in the different cities. And also when you think about like, um, I think, yeah, I mean, we're going to get into it as well, um, a little bit. But like all the stuff that we're basically learning now, there's like such a huge open source community on the Internet, all the materials pretty much there for you to learn and stuff. So I think I'm just benefiting a lot of like other people's work that they put on the Internet. So I think that's, it's going to be my favorite piece of tech.

Speaker A: Brilliant. I mean, where would we be without it?

Speaker B: Yeah,

Speaker A: so we met in a kind of different way, didn't we, Nicholas? So, yeah, I received a really Lovely message on LinkedIn from someone called Eric Levy. And they said, uh, um, they'd recently stumbled across the podcast and loved it, uh, that they were studying data science. And I was like, oh my God, I love it. Thank you so much. It means, so it means the world to me. And I, and I asked him how he, um, how he came across the podcast and he said that, um, you are one of his instructors at, um, General assembly, showed it to them. And I was like, really? Because you knew Mark, didn't you? That came on. Yeah, um, Mark Stevenson. Yeah. So you shared my podcast with the people that you were teaching at General assembly, which is amazing for me. I love that. So can you kind of talk me through, firstly your career?

Speaker B: Yeah.

Speaker A: Um, and then we'll talk about General Assembly a little bit.

Speaker B: Yeah. I mean, also, first of all, I think, uh, I think on that podcast, by the way. Uh, I think, yeah, um, there was. You didn't just have one person that I know of on the podcast. You already had General assembly instructor before. Yes, yeah, but I think at that time you didn't, you weren't aware that Fareed was also doing that. Because I think he's doing like consulting for the military and stuff like that. Like, uh, pretty crazy stuff. Um, but yeah, so yeah, Mark sent me that podcast and because at ga, um, as well, obviously a lot of people are very interested in how to get hired and stuff. And I thought it was a really cool podcast to share. So those, um, that was really good. As for me, um, I originally started, studied ah, economics, which I didn't like at all. Like I basically, I wouldn't say that's even my background. I think the main thing I took away from, from that degree was the stats classes at the end, which I really enjoyed. And I was working sort of in sports management a little bit, um, on the side. And I was kind of thinking like, how can I combine sort of stats with sports? And at the time sports data analytics, um, was growing as, as a field and I thought, oh, that would be really cool if I could work like in basketball, football analytics. Um, and so I was like, okay, what do I have to do for that? And I found like a data science masters in Barcelona at the time. And so I applied and I, and luckily I got in and finished um, off the course by doing uh, my master thesis with FC Barcelona, which was sort of a dream at the time.

Speaker A: Yeah.

Speaker B: And then following that I found a job at the Zone. So that's where I met Mark, where we basically did like mostly customer analytics. So we're looking at how do people interact with um, the app, which is basically like a broad, like they label themselves the Netflix of sports. Um, so yeah, we were basically just looking at how do customers interact with the app, ah, what makes them stick with the app, what makes them maybe leave the app, um, stuff like that. Um, and then I um, decided to move on to basically football analytics company which was also really, really cool. Um, um, so there we basically just try to predict, you know, um, how likely it is that you know, the results of football games, um, plus also trying to evaluate players and stuff. Basically what I've, you know, kind of, you know, always wanted to do. Um, and then I kind of like this was another thing that I always wanted to do, basically go um, the freelance route or work for myself. So I started, I moved to Malaga here in the south of Spain and started, um, um, an agency for data science products or data products. Um, and also I taught myself a little bit of web development, which is also what we offer.

Speaker A: Okay, amazing. So just going to the football stats thing, uh, how often were you. Right,

Speaker B: so it doesn't exactly work like that. What you would do is you would create odds, right. You would basically say, if we have, like, if these teams would play against each other a thousand times, how often do we think this team would win? How often do we think that team would win? Um, so that was kind of the. That was kind of the job. And I would say that the model was really, really accurate. Although I can't take a lot of credit for that because the company has been existing a lot of times. But, yeah, it was basically already a really, really good model, um, that just needed to be tweaked.

Speaker A: And yeah, there's lots of things like that. I mean, I mean, data science is literally everywhere, so people listening. If you have got a passion, you're passionate about something in life, it could be anything. Ah, we spoke to somebody who used to work at a bakery, so now they're doing data science for bakers. Um, you know, you've got your football, you've got like literally data science, absolutely anything. So if you've got a passion for it, do some projects on the things that you've got a passion for. Um, not only will they probably work better, uh, but you will be able to talk about them so well, because you understand it, you've got the passion behind it. So that's a real top tip there for you.

Speaker B: Yeah, I. Sorry, I really, really agree on that. And I see it again and again as well with General Assembly. Um, a lot of times students would come in and be like, oh, what does an employer want to see? When, you know, I build like a capstone project. I remember this, um, this one student and she, she was absolutely brilliant. But she was at the beginning thinking sort of along the same lines and was thinking about projects that she wasn't really that interested in. I think then she couldn't find the data for those projects and she decided to pivot into a project where, um, I think she was basically doing like, um, nlp, so Natural Language Processing, where she was looking at Star Trek dialogues and trying to predict which characters said what. And it was a really, really good model. And the presentation, everything was so great. Yeah, um, yeah, it's one of the best projects I think I've seen. So it was really cool.

Speaker A: Yeah, really, really great. So, um, talk to Me a little bit about general assembly. Obviously there are other suppliers like boot camps and stuff but talk to me. What do you do? What, what do you offer?

Speaker B: Yeah. So um, in general the idea. So I'm teaching the data science and as well the software engineering um courses there's immersive courses which are uh, kind of like the main bread and butter that everybody knows about. Bootcamps basically just 12 weeks, um, every day from nine to five or nine to six actually. Um, and what we do is mainly, I think the students are mainly like mid-20s or later, um, usually that already had a career um, that they didn't really enjoy and they look to, they want to pivot. Um, so we basically try to give them, you know, I would also more say it's applied. Data science is not the same necessary that you would learn in uni where you have to do a lot of theory. Yeah. Instead we're basically just trying to prepare you that when you come out after those 12 weeks um, that you're going to be at the level of like a junior data scientist that can apply everything that has seen like uh, most of the possible applications of data science in the, in the real world.

Speaker A: So what kind of. So it's, it's, it's a full time thing and it's quite intense, isn't it? I've heard. Yeah, quite intense. So not to muck about, you're going in there to learn and you're going to learn hard.

Speaker B: Yeah, I think, yeah. In general there's um, like we want you to struggle a little bit. Um, it's just, I think uh, I mean we don't get off on it. It's not something that we enjoy. But in general um, you should, you know there will be later on, um, in the course there will be moments of despair where you just, just too much and all that. Um, but it's okay because just as long as you stick with it and we have exposed to some material you're going to, you know, maybe after the course revisited and then um, then it will all click. But yeah, it's pretty intense. Um, but we're obviously that the whole, the whole time to support.

Speaker A: Yeah. So throughout the say 12 weeks. Yeah, throughout the 12 weeks. Obviously there's lots of different modules and I know that you don't do all of them but can you give me an overview of what kind of things are covered?

Speaker B: Yeah. So we basically start off so this, the bootcamp is just in Python. Um, so we're starting off first with Python fundamentals. Um, like A general introduction into programming. Um, then we go into the libraries that are mainly used. So the whole thing like Numpy Pandas, um, matplotlib, um, all of those. So that you basically can do um, exploratory data analysis. Um, from there on you basically move on to um, modeling. We start with supervised learning. So lots of different, um, classifications, regression algorithms. Um, there's also some stuff in between where you start, um, working with external APIs, um, databases. And then there's. So where I come in is basically unsupervised learning. So clustering algorithms, which is basically what I've done a lot as well at the Zone. Um, and then after that we have a week in stats and then the end of the code or like that, that's now week eight or nine and then the last three weeks is um, a lot of also like cloud computing time series. Um, there's a very, very short introduction into neural networks, um, stuff like that.

Speaker A: So like a really high overview of pretty much everything.

Speaker B: Exactly, yeah. And you basically, you will, when you come out, you will be basically have run through, um, you know, I don't know, you have, you will have fitted like, uh, I don't know, random forest or whatever on a data set. So you will always have at least some notebooks where. Or even neural networks. Right. You will have fitted a neural network to um, some problems. So you will always be able to go back and look through your notes even though, you know, some of the stuff might not exactly stick in the first moment. Just because like you say, it's the overview of everything you can basically learn, um, within 12 weeks.

Speaker A: So how many kind of, how many projects are, ah, do you do in the course?

Speaker B: Yeah, so there's four, um, projects and a capstone project. So the capstone project is the big one, the one that um, is also the one you're supposed to show to your uh, employers or future employers to show what you've done. Um, sort of combine everything that you've seen up until then the course. Um, and then there's some, some smaller stuff that is just there to um, basically make you practice what you've learned during the. During the course. Sure.

Speaker A: So I'm thinking that, you know, um, GitHub, for example, a lot of people, a lot of employers would like to have a look at your GitHub to see how you work, how your brain works, is there. Do you go. Do you cover GitHub with people and show them how to. So all of their projects are on GitHub?

Speaker B: Yeah, exactly. So we basically, that is in the first week we cover um, Git, or at least what they need to know to push their code. And also everything that we do with them is um, like basically all the lesson materials are also shared on GitHub, so they have to, they have to pull it down to the machine.

Speaker A: Okay, good, you've got your answer to my questions very well. Um, and so is this course for somebody who's got no idea, has never done coding, has never really done anything, Data science think that they like the look of it, um, and want to get into it, or is it for somebody who's a little bit more knowledgeable?

Speaker B: Yeah, no, definitely. I mean it's definitely for, for everybody. Um, we have people with uh, like with very much backgrounds. That would not be necessarily what you think of when you think data science, like anthropology or stuff like that, or literature, um, like English literature or something like that. Right. You know, um, you have people from, with, with any kind of background. And it's really also geared towards people, you know, that, um, that are sort of starting from zero. I would say however that if you sign up for a bootcamp, and again also all the stuff that I'm talking about, it's not just general assembly, it's probably also all the other bootcamps.

Speaker A: Yeah.

Speaker B: I think the best thing to do is to at least try to prepare as, as good as you can. Like if, you know, the course is going to be in Python, you know, do some course or do some stuff before on Python. Because the idea is sort of that the, you know, the course will move along and the easier it is for you at the beginning to keep track of what we're doing, um, the less you will struggle later on. Right. So the people that are already struggling at the beginning a lot just with understanding Python, um, when then the steps part comes, you know, together with it, it will be a lot harder for them to, you know, to compute, uh, everything of that. Right. So I would say be well prepared. You don't have to like, if you decide now to do a bootcamp and you have like three or four weeks before it starts, I would just say use the time to, to get up and running, at least with the language you're going to use.

Speaker A: Yeah, absolutely. So either have a look at Python or if you've got other, another language like Larva. Ah, and Python are really similar, aren't they?

Speaker B: Um, I mean, yeah, you can use both for, for data science. Um, but R is basically mainly so R's language that is, that was made for Stats. While Python is kind of like a universal, uh, this is very nice saying. There's basically, um, I think Python, Python is the second best language for every program, for every problem you can encounter. Right. There's, there's like one language that will be very specific to that and then Python you can basically do.

Speaker A: Great. I've learned something today. Thanks very much. Thank you. Thank you. Um, so if somebody comes to do uh, a data science boot camp. Sorry, Jimmy's just digging around and he's found the wall. So now the room's covered in wall. Um, sorry. If, um, if somebody comes to a data science boot camp, how do you support them or set them up for getting their first role? Because junior data science roles, graduate roles, they are very few and far between. It seems at the moment there are a lot more people than roles.

Speaker B: Yeah. Um, so, yeah, first on that, um, I think that's true. Like, um, I think it's probably harder for a junior to find a job right now. I also think it's. When I, when I look back, I was just talking about this yesterday in class, um, like when I got my first job at the Zone, right. It was called Data, uh, scientists. And I think, um, the data science role itself sort of branches out into lots of other roles today. Back then people basically didn't know what they wanted when they hired a data scientist. Now you basically have, I don't know, data analysts, machine learning engineer, um, like research analyst, business intelligence analysts, whatever. Right. You have, you have so many new roles and basically it means that people coming into a job should probably, you know, try to specify on um, some role that they're actually interested in. Right. So it's not like if you're looking for data science role, they're probably fewer, but there's overall probably more roles in like the data space.

Speaker A: That's actually a really good point. Yeah, Engineering as well. There's, there's more jobs than there are people. Um, so if you're, if you're more on the technical side of it, then it's worth the thought getting into engineering. It's uh, exactly as well.

Speaker B: Yes, exactly. So data engineer roles for example. Right. Have just been booming, I think. But that also. Yeah, like you say, that doesn't make it necessarily easier for the, for the junior roles. Um, at GA at least, or assembly at least. Um, what we do is there's an outcomes team. So basically they prepare you if you meet them on the first day of the course and then they will be there, um, I think every two weeks. Um, you Have a session with them that basically trains you on like, how to network, how to build your cv. Um, they will also coach you through, um, I think interview stages and stuff. You can always reach out to them and ask them, like, what do you think of this? Um, and then I think there's also. No, I think there's also stuff, uh, like ama. So because we have been running the data science, um, course for quite some time, there's obviously alumni that come back, um, and tell them about, you know, how their, how their job search was and how they're now finding their role and stuff like that.

Speaker A: Brilliant. That's really good. So your support from really from the beginning to the end, uh, and look, so you, you spoke to Mark about. Does he have any jobs? Uh, so I know that you do reach out to people to find out if there are roles for, for the people in your class, which is great. Um, okay, so thanks for that. So what would your best piece of advice be for somebody just starting out?

Speaker B: I think, I think just sticking with it. Like, um, I think there's one, there's one kind of problem that data science has, which is this article that came out about it being the sexiest job in the 21st century. So a lot of people that are not really interested in learning it are probably just starting it for the, for the money. Um, potentially. I think as long as you, uh, as long as you're like, um, I think you should before, you should probably look at it and see like, is this something you can see yourself doing? Because it's a, it's a pretty tough road ahead of you. Yeah, um, um, as long as you're, as long as you think like, oh, okay, I can work in front of my computer eight hours a day, you know, 40 hours a week. Um, and I can be interested in finding or like drawing insights from data. Then, then I, then I'm 100% sure you can do it. Like, I haven't had a student yet that was motiv and that couldn't go through the 12 weeks and afterwards come out and be, um, or like, have learned a lot, a lot, a lot that is going to get them employed at some point. Um, but I think there's also people that, you know, just look at the money and then it's going to be tough because if the main motivator is that you will probably not work on the weekends and try to figure out like, oh, how does this algorithm actually work? You basically just there to say, oh, I want this, I want this to work so that I can make the money, basically. So I think it's very important, like, if you, if you are motivated, you think this could be a role for you. And the same thing goes for, I mean, it doesn't have to be data, right? If you don't find data that interesting, there's other tech roles, like for example, software engineering, if you're interested in building websites, all of that stuff. The learning is going to be really hard. It's really, it is really difficult. Um, but I think everybody can do it. So I haven't found a student yet that, you know, actually was motivated and was working on it and then didn't make progress. So I think, I think just sticking with it and trusting through it. And there's a lot of people that are kind of after like two or three weeks already demotivated and thinking it's not for me, I'm not smart enough. Um, but it's just if you think about how long it takes you to become a proper developer or data scientist, the first two or three weeks are just not predictive at all.

Speaker A: Yeah, yeah, absolutely. Uh, a question popped into my, into my mind about, um, General Assembly. Do you teach everybody the same way? Now I'm thinking of people who are neurodivergent here. Uh, like my daughter, for example. She cannot learn things. If you talk to her about it. You, you know, you have to, she has to do the practical. And I know that you're going to do both of those things, but. Yeah, so do you teach everybody the same or is it, uh, or people who are neurodivergent, for example, able to come along?

Speaker B: Yeah, yeah, it's a really good question. So in general, like you said, we have, um, Usually like the morning is for um, like lecture time. So I'm just going to be talking to you about, I don't know, certain algorithms, how they work. And then in the afternoon you have time to practice, um, by yourself. There's, there are people that are a little bit faster at the beginning or slower. So we do have material that basically accommodates both. Or like, if you, um, if you are rushing through it, then we have probably additional stretch material you can look at. Um, on the other hand, students that are, that are struggling as much are just going to, uh, that are struggling a lot are just going to get a bit more personal support. Right. Then you're going to be like, you know, others might already be further ahead, but we're just going to go into a breakout room here and we're going to talk this through and then you know, so, so you definitely, you definitely spend more time on some students than, than others. There's not necessarily different material for, um, divergent students, but.

Speaker A: Yeah, yeah, no, no, that makes sense. That makes sense. Especially having the different ways of learning in one day. Um, I'm just thinking back to school. It just, when you were just talked at and talked at and talked at and talked at, it just doesn't work.

Speaker B: It doesn't work for programming. Um, I think there's. Yeah, generally there's always a distinction between self taught and not self taught. But to me, like, everybody is pretty much self taught because nobody can learn by looking at somebody else code. Like your brain actually has to look at like, oh, what do I actually write in this line so that this works that you know, it can't be told to you. So I think, uh, everybody's pretty much, yeah, self taught.

Speaker A: Brilliant. Great answer. Thank you very much. So you've got a master's degree.

Speaker B: Yeah.

Speaker A: Uh, do you think it's important for people to go down that route? Probably not. As you're working at General Assembly.

Speaker B: I think it depends. Yeah. So, so for, depends on the role you want. Um, so I personally would say no, but for me, for example, I needed, you know, the master's thesis with FC Barcelona was really important for me to then find a job after. Like, I just learned so much by working on, on an actual project that I was passionate about.

Speaker A: Yeah.

Speaker B: Um, which came sort of with the masters, but the masters was pretty theoretical and it was, um, basically helping people to potentially pursue a PhD afterwards, which then is basically just about proving theorems and you know, doing proper research and stuff like that, which is like a minority of people that are actually doing that today. In data science, I would say most of it's probably more applied data science. So, um, but there are people, for example, that do boot camps as well and then decide, oh, this hasn't, you know, this wasn't deep enough for me. Like, it didn't go to the core of what stats is. So they then decided to maybe do a master's afterwards or like specialize in that. So it can also be an entry point. Um, I would say in general, if you're, if you're more interested in the applied side, um, bootcamps are probably better. Um, if you do want to go into something like research, um, engineering or machine learning researcher, uh, or something like that, then a master's is probably better. But boot camps could still be a starting point.

Speaker A: Yeah, definitely. Um, and like, is there, do you do like, is There just one bootcamp that you do for data science or do you do a return as one so it goes into a little bit more depth?

Speaker B: No. So yeah, there's, there's one boot camp or one, one corpus of material basically. Right. That we, that we teach. Um, there are some things where. So for example for software engineering we've done it. And for data science, I think that they're planning on potentially doing that as well where instead of 12 weeks, it's going to be 24 weeks.

Speaker A: Okay.

Speaker B: And you just have, but you still have the same material. You just have more time in between to um, to learn it. So this might also be something when you're asking like, you know, people learning differently. I think for example, spacing, um, is very, very nice in learning. Right. So basically having a course or like a lecture and then having like one or two days to processes and process it and work on your own and then have another lecture.

Speaker A: Yeah.

Speaker B: Um, but yeah, the general idea is, yeah, we have, we have one um, one boot camp material.

Speaker A: And how do you touch on ethics or uh, bias in data?

Speaker B: Yeah, so there's, there's definitely discussions about it. Like we don't, we don't teach it in that way, but it automatically, um, it automatically comes up. Um, so there's basically all the communication happens on Slack. So a lot of people, like every morning before we go through, um, every morning before we start with the lecture, um, we have a stand up where everybody talks about like how, you know, how yesterday went for them and what they're blocked with, but then also whether they want to share anything else. And students usually start to pick up on things in data while they do the bootcamp. Right. They start going to like meetups and they also start reading articles on Medium. And then we also start, you know, they often mention things like, oh, this is interesting to think about. Like here you can use, here you can see like how an algorithm might have been racist in um, I don't know. Ah, right. And then, and then you do start talking about it and you start uh, sharing it on Slack. But we don't have particular material around it. But it, I think it pops up.

Speaker A: It's just going to come up, isn't it? It's just one of those things. Yeah. No, that's great, that's great. How do you think that we can stop bias in data and algorithms?

Speaker B: It's um, it's a really tough one. Like I think you basically have to have people that are actually working on the algorithms and mainly also designing the Data. Like there's a lot of bias when you're, when you're actually um, when you're gathering the data that you don't think about it and people don't, I don't think people do it intentionally, but rather they start with themselves and they don't really think about like, you know, other people that might have different challenges and when they, you know, when they get the data, they might make that mistake. So it's really tough to say. I think you basically have to have like a very mixed, um, group of researchers that actually gather the data for your algorithm that think of all the different, uh, things that could impact, um, by the, or that could create bias in the data. But it's a, it's a tough problem to solve.

Speaker A: It is really tough. You know, during the, during the time on podcast, I've, I've, it's really been a topic that I care about and so I've asked the questions about how can, you know, how can we. Historically, data is massively biased, right? Massively biased because it's always started data was just uh, white man, right? And then I was like, how can you change that historic data? Because, and I've learned now that you can cut it off at one point or you can do two different, or you can run two different models or, and then you can retrain the model. So if anyone wants to know any of that, go and listen to all of the other podcasts. Podcasts. There's only, there's only like 48 of them. It's fine. Um, so, okay, what do you think data science is going to evolve to? Do you think it's gonna get bigger or do you think it's gonna get smaller?

Speaker B: That's, that's a tough thing. Um, I think the, the trend that we currently see with like more um, specific roles will probably continue. I think. We're not, we're not finished there. I think also personally, um, that engineering is just going to become a lot more important like writing proper code and not just working your one off models where you draw your data from like a CSV file and then you uh, you know, send some graphs around in an email. I think those, those positions will become less and less. You will have to be able to work in a data pipeline, um, and you will have to be able to write code that can actually go into uh, a proper code base or data scientists will, will get closer to data and uh, data engineers, so they will have to write code. Other than that, I'm not, I'm not too sure. Like right now, when I like the consulting that I do or the um, the. Yeah. The data products that I build. Right. Is mostly for a specific set of customers. So, um, so I can't really talk about like what, you know, data science goes for zone of a big.

Speaker A: I suppose there's. People are always going to have problems. There's always going to be some, a problem for, for you to solve. Uh, no matter how much automation there is, there's going to be a new problem that's going to need solving and you're going to have to solve that for them.

Speaker B: And I mean also, I mean it's a, it's a smaller subset but there's still people that have to code up those automation tools and have to, you know, building those and stuff like. Yeah. So I think tech is not going anywhere and data science is not going anywhere, but it's.

Speaker A: I'll bless you. So before we wrap up, Nicholas, have you got anything that you would like to plug or talk about?

Speaker B: No, I know I came completely without anything to sell. Um, that's great.

Speaker A: That's fine. I mean you can sell my coaching service.

Speaker B: One thing, not, not on me, um, just in general because we talked a lot about data science.

Speaker A: Yeah.

Speaker B: And I was, I was saying this earlier a little bit when I was talking about the, like the sexiest job. Um, I think, yeah. Just, just check out as well, software engineering. I'm really, I'm really passionate about that as well and I can see that when I, when I teach that there's. You can just tell from like some students that, you know, they might really prefer the software engineering side of the data science. Right. Just, you know, actually inform yourself before you make um, before you make that decision.

Speaker A: Yeah, I think so. To go in with an open mind and have a look at the different roles. You know, you can even reach out to me and I can go over what the roles are. You know, I've got so much like job specs that I can blank out, so nobody knows who it was or write. Write something similar and send you the difference in job specs and have a chat with you about that. That's not a problem. M. Um, yeah, I think that's good.

Speaker B: And I think you can also reverse engineer the process. You can basically look at, oh, this is the kind of job, like maybe this is the kind of company I want to work at. They have this job. This looks really cool. What are they actually looking for? And technology that I have to learn and m. Basically look through like which path of career should I take to actually end up in that sort of world?

Speaker A: That is a really good shout. That's really good. Right? Nicholas, thank you so much.

Speaker B: Thank you.

Speaker A: A real pleasure. And I definitely recorded this one. I really appreciate you coming back. Thank you so much for joining me.

Speaker B: Thank you, Wendy.

Speaker A: Take care.

Speaker B: Uh, la.