#172 How to make Developer Friends When You Don't Live in Silicon Valley, with Iraqi Engineer Code;Life
01:29:27 Share

Welcome back to the Free Code Camp podcast. I'm Quincy Larson, teacher and founder of freecodecamp.org. And today I'm talking with software engineer and live coding streamer, Codelife, aka CL. For those of you watching the video version of this interview, CL lives in Iraq and she uses a 3D VTuber avatar to protect her identity. CL, welcome to the podcast. Thank you so much for having me.

Yeah, and I want to dive straight into talking about machine learning because that is an area that you've spent a lot of time working. You were fine-tuning language models for low-resource languages back in grad school, back in 2018. What is a low-resource language, and how has the fine-tuning process changed since then? Right. So, yes, there has been a lot of progress in...

working with languages and computers. And back in 2017, 18,

The tools existing were different, especially for languages that are high resource like the English language and some other European languages and maybe even a language like Chinese that have a lot of people on the Internet creating content for it. But low resource language is a language that even maybe there are many books, many books

non-digital content available, but something like optical character recognition, OCR, has not been to a level that is perfected for it to be used at a mass scale to digitize it. There isn't something like

Thousands or tens of thousands of Wikipedia pages for it. If you see the Wikipedia chart of number of pages for languages, you can see that languages like English are at the top and other. There's like you can see a long tail of other languages on the chart. And most likely these are low resource languages here.

Yeah. So just to recap, so English is by far the most well-represented language on the internet. Like about half of Wikipedia's articles or at least half of the readership of Wikipedia articles is English articles. And then you've got your German, your French, your Chinese, your Japanese, like other languages that 100 million plus people speak.

Uh, but you know, it very quickly goes down until you're talking about, uh, languages that, that may be very literary in nature, but it's been like, like printing on books, like, like,

places where maybe the penetration of computers hasn't been as, you know, substantial as it has been in the English speaking world, for example. Yes. Yeah. So we talked about what a low resource language is and that's the general idea and creating tools, even not having enough tools, like we can call them a poor language digitally. So,

Creating tools for those languages was something that I was interested in during my master's. And back then, I was using something called an Ngram language model, which is like basically writing on a cave wall. Even though it's something that is being used now as a base, it's...

That, compared to now, where I was able to test existing models for text-to-speech or speech-to-text or even fine-tuning, which is what we're talking about, taking a model that someone in a big tech company worked on, used the whole internet on, and I'm someone giving it a portion of my data, whether it's text, whether it's audio, and...

running the model multiple times using GPU, making it familiar with the content that I'm giving it and testing it for my language. And this can be done

your language doesn't have to be Latin-based. It can be something like Korean that uses Hangul or any other languages that are available. So the interesting thing recently is that, yes, tools are not being made with multiple languages in mind, but people like me who are interested in that can utilize those existing tools that exist

takes probably, I don't know, millions of hours to train and build on that. It may not be something that can be done easily, but the progress from back then to now is very, very high. And the amount of data that is needed is also going lower. Yes. Okay. So...

Why is it that these low resource languages where there's just a whole lot less text that can be fed into the model...

Why is it so important to have so much text? Like, for example, English Wikipedia, Reddit, all the blog posts that everybody wrote in 2010. You know, like those data sets. Why does that make the model perform so much better by virtue of having so much text? And why is it that when you have a much more limited amount of text to work with and to feed into the model, do the results from that model not?

Why are they not as good? That's a great question. We can even boil it down to something like if I ask you to read a paragraph and then tell me about...

Let's say, what is astrophysics? You will try and repeat the parts that are in the paragraph. But if I tell you, here, read this book on physics and astrophysics, then you have a lot more to draw from. And what a language model, depending on the type, definitely changes, especially with large language models. But it changes.

to create connections between those, whether we call them words or we call them tokens or we call them subwords. These are things that are being used. It will represent it in a

3D environment, like word embeddings, and it creates connections between them. And an example that we have seen a lot on the internet, it would be like, if you say capital, and I think I spoiled it, it's supposed to be, if you say Japan, Tokyo, and then you remove, say, Japan, Tokyo,

minus capital, plus, like, let's say, France, it should give you Paris because it created those connections between the words. It knows what is the relationship between those. So the more text you give it, the more it can mix it and get, I'm doing quote-unquote, understanding of the language. And it is the same for images. We have seen how...

the amount of progress from the, um, dreamlike, uh, hallucinations that were made, uh, I will say again, uh, before COVID BC. Um, yeah, like the Google dream project or what was that? Where like everything had like lots of eyeballs in it, like animals in clouds and on mountains. Um, and now, um,

You can see the difference because something that I was very interested in back in the day, I think it was like 2015 or 16, I was watching Dr. Faye Faley talk about her labeling millions of pictures

pictures to create a data set. So data sets are the oil, sadly, of language models. And there is a saying that says trash in, trash out. You give it bad data, it will give you bad results. If you give it a small amount of data, it will create weird connections that then become its own type of hallucination. So it's a really interesting field, especially for like

of like the English language is trying to optimize, optimize, optimize. And we're out here saying, oh, you created something. Let me utilize it for my language and have like our people use it. Yes. Yeah. Yeah. What are some examples of low resource languages where...

there's a big effort to try to like digitize more stuff. And maybe like if you can think of some low resource languages that like millions of people speak this language, but it's just not that well represented on the internet. Yes, definitely. Actually, there is a...

like an organization style of people that are called Masakane. Masakane? Yes, they are working for the African languages. I think one example is Zulu. I'm not sure, but... Like, I know one of them is Zulu, but I'm not sure if Swahili is, but there are people from, like, all over Africa working as a group to create tools for those languages that are not the thing that...

like funding probably goes to. Yeah. Yeah. And just Swahili is a good example. We actually have localized a lot of free coat camping to Swahili because we have a Swahili speaker on our team.

And it's a language that like 80 million people speak, but it's just not represented on the Internet. So these efforts with models like to try to get Google Translate to work really well for Swahili, that is dependent on feeding a whole lot of tokens, a whole lot of Swahili words that are...

spoken by humans and stuff that, that have like actual meaning to them. Right. When you just generate a random string of words and throw that into the model, that's not going to make the model better. That'll make the model worse. You want more signal and less noise. So yeah. Can you describe the fine tuning process? Like what exactly is fine tuning for people that are, that haven't heard that term before? Um, um,

So there are different ways of fine-tuning different, I will say, available models. Like a very practical example would be I'm interested in, let's say, summarization. I go on a place of, like, resources like Hugging Face, which is a place that I use quite a lot, actually, where you could download available models depending on how much, like, resources

resources you have like on your computer. And you open that, you go into whether it is their documentation or a website that's available, or even most of them are on GitHub. And I go and look at what choices they have for fine tuning. I have like personally tested, I can say over, I'll say over 10,

that is for text to speech. And I'll go look at the part that says, here's how you fine-tune our model. I try and run it. There have been times where it doesn't work because it has not been

created for things like Unicode or different scripts. So I modify the code that's there, but I will say that a lot of projects and a lot of models tell you how to fine-tune their model. They say, we're using something like...

what type of data set we expect. So you prepare the data. That's like the main part. If you have the data and it's prepared based on the way they want, for me, I think more than 50% of my efforts has been to prepare the data, process the data that I have. And even then, a good percentage of the time I have

use data that's available online. An example would be Common Voices that has both text and audio data. And you can download that. You can process this however you want. I have used Python to modify the data the way I wanted, data frames. And then you feed it onto the style of fine-tuning that this model expects.

But there have been times where a model I find interesting does not have any fine tuning related documentation. So I go and search and see who has done anything about it. I won't say that I'm an expert in like fine tuning things from scratch, but I have seen like people blog about it, which is amazing sharing resources, other people making videos about it. I have also done that. I have tested other people's code, have it not worked for me,

ask something like Cloud Sonnet saying, oh, can I, I don't understand this part of the code because it's too advanced for me. And then having it help me understand it and then modify it the way that I want it. So it's a process that is very, very custom based on the language you're using, based on the model you've selected, based on the task that you want to do. Yes. Wow.

So there are a lot of instances where, like, a particular model won't have documentation, won't have some sort of, like, fine-tuning API, and you have to just figure out how to do it yourself. Yeah, it's pretty cool. I definitely want to compliment your audacity in, like, I...

This is not like a documented feature. I'm going to figure out a way to bust a hole through this and get my data into the model. You don't always succeed, but when you do, it's amazing. Yeah, awesome. And also, it's cool that you're working with so many different models because a lot of people just use the same handful of foundation models. I'll use GPT-4 or I'll use Claude or I'll use some...

I'm sure there are specialized models for accomplishing specific tasks, but what you're talking about, like speech-to-text, text-to-speech, you can use machine learning for so many things other than just asking questions in a chat interface, which is, I think, what a lot of people... That's the primary way to use it. Yeah, generation has been the thing that people see the most, but I prefer specific tools, right?

Like spell checking is like very specific, speech to text, transcription. These are like specific tasks and tools because then you can, in a way, perfect it for that task. Even though like something like Whisper, the open AI model that is for audio to text, also does translation. So they have found out based on the paper that I've read.

it does a better job of doing both tasks instead of doing it one separately. I'm like, Hmm, I have a hard time believing that even though that's what they have tested. Yeah. Yeah. I also, it doesn't make intuitive sense that it would be easier to both translate and, uh, transcribe, transcribe, but you know, uh,

Good for them. Yeah, yeah. Again, I am skeptical of some of the... You shouldn't just take everything as gospel when you're looking at documents that have been published. Everybody has this publish or perish thing. And opening eyes, their future stock price is dependent on being able to break through and being the first to do X, Y, Z and stuff like that. So there's a lot of competitive pressure. I wouldn't be surprised if some things...

where, oh, actually, you know, the p-value was very low for this or something like, you know, the kind of things that usually come up in retracted papers. I wouldn't be surprised. I wouldn't believe every headline you read. But it's good that you're being skeptical about these things and that you're kind of digging in and seeing for yourself. Yeah. I definitely want to applaud that. I want to learn a little bit more about your background as a dev. Before we dive into that, I just want to thank...

The many, many people. We've got more than 11,384 kind folks out there who support Free Code Camp. You all make the Free Code Camp podcast possible. You make the entire community possible. The platform where we are publishing new interactive curricula. The YouTube channel where we're publishing full-length courses. The publication where we're publishing full-length books on different programming concepts. Everything.

Thanks to the community. And also I want to thank Wix studio who've created a grant. Wix studio provides developer tools to rapidly build websites with everything out of the box, then extend, replace and break boundaries with code. Learn more at wixstudio.com. And again, if you want to support the free code camp community directly, go to donate.freecodecamp.org. So CL, you grew up in Iraq and it's certainly in America, uh,

We think of Iraq as that place that we invaded like twice. And then, of course, there have been a lengthy history of war with Iran, the neighboring Persian state. And I wanted to kind of just...

get a feel for what it was like growing up in Iraq. And maybe if you can just put some perspective and some, some color on what I think are relatively like, we just have like, probably if you grew up during either of those wars, you just have all these like newspaper images and you're like, okay, that's Iraq. But I'm sure Iraq is a lot more than what was captured, you know, 20 years ago in newspapers. Uh, can you, can you talk about what, what your child was like there? Uh,

Sure. Thank you for asking and for being curious. So I will say that my experience is my experience, and I'm sure other people had different experiences living in my country.

I grew up probably being sheltered and not having as much information about the struggles that my family went through. Like now asking like my mom questions and asking my siblings questions of like back in the 90s where like when I was a baby and don't remember anything. It was a time of much, much struggle. We were like fresh out of a few wars and not having like...

enough to feed the family, um, working extra things, working for people. These are things that, um, my parents and my family went through and I just grew up thinking everything is great. Um, I have a lovely family, I have food to eat and I did not find any struggles whatsoever. I had like, um, siblings to look up to and, um, to bother to ask questions. Um, so in that sense, um,

Me learning about the past is something recent. But I'm talking about the 90s. I have not come to the 2000s yet. Yeah. So, yeah, I guess if I talk about the rest of my childhood, I...

Also grew up having access to DVDs and like entertainment that was mostly from the U.S. So I just grew up listening to the English language, learning from the older siblings, them knowing that it is something that I'm going to need in the future and them having like giving me resources. An example would be I doubt you know anything about this, but have you heard of Magic English?

No. It was a learning English program that was using... I'm guessing it's made by Disney. I hope it is because it was using everything Disney and just like teaching you. Yeah. Like teaching you things, repeating the parts. So it was something that I had memorized when I was a kid. And it was something like a game for me, but the...

But the family member who bought it for me was like, she's going to learn English much faster through things like this. I grew up memorizing, again, Disney movies, Mulan. Shrek is not Disney. But things like that. Yeah. So me having that exposure. That's so interesting. Sorry to interrupt, but I just want to observe. You haven't spent a great deal of time in English-speaking countries. Have you lived abroad?

I traveled to the United States for four to five weeks. Four to five weeks. That's it. Your English is excellent. And I used to be an English teacher. That was like a big... I was an English teacher and a school director before I learned programming and became a developer. And I would say your English is like...

Like you would be the best student in our school basically with your English. So, and I was teaching adult learners, you know, in their twenties and thirties and forties. So yeah. So magic English definitely helped your passion for memorizing lines from Shrek and things like that. Definitely seems to have played a part. Your pronunciation sounds very North American. Yeah.

I remember when I was like, I don't know, maybe in middle school deciding I'm going to have an American accent, not British, because that was our curriculum. Our curriculum was British. So I was like, I don't know why I think American is like something I like.

I really don't know why. Yeah. Well, so you're learning English. That's great. And what kind of doors does learning English open up for you? Like my understanding is eventually you got a computer and you got Internet access. Can you take us back to that time? Sure. Sure.

Yeah, I don't think I remember the first time seeing a computer because I feel like it was something that my brothers had, something in the background that I, little by little, I was definitely encouraged to use it. There's a thing called, again, a tool from the 90s called Child's Play. It was basically paint, but a thousand times better. And

It was something that I would just play with, show the neighborhood kids, like, look, I made something in this. So it was before the Internet. I was just using it for fun. I remember also having a program that was a keyboard typing program that you were driving. And each time you made a mistake, a bug would land on your shoe.

It was really fun. So I had things to use. And then when we had like dial up internet, I'm guessing, I don't know, I can't say exact time, but let's say around 2005. I like was like going on Wikipedia, looking at the movies that I watched.

I would like when I found YouTube, I think my world just turned upside down. Just like watching. Like, how do I do this? How do I do a handstand? How do I from then? And then I just started following people, watching vlogs. And like at a point when I was probably in high school, like I would look at a topic,

in my science book, and then I would look for a video about it. Or I used to be really interested in space, watching lectures on it, stuff like that. So I think not having the...

of thinking English is just something natural, I would not have had the freedom to explore stuff, even creating PowerPoints and having a picture jump around. Just being familiar with computers, which I do feel like the current... Is it called the Al-Fajan?

Like after Gen Z. I feel like they don't have... Generation Alpha, I guess. Thank you, Generation Alpha. I'm not sure how it's called. I think you're right. I don't think they have the same familiarity of using computers and not like phones and tablets. Yeah. So you got to...

do computing when it was hard. And of course, the people in the 70s had it really hard. They had to literally assemble their machines following the manual and stuff like that. And they had to literally type their programs into the computer to be able to run them. No, no. It was all UI for me. Yeah, yeah. So similarly, I didn't get the internet as a kid. But you were making heavy use of the English internet. Can you compare and contrast like

the Arabic language internet at that time? Were there as many resources in Arabic? That's a really good question. And I don't think I can even answer you. I don't think I have... Sometimes when I go on a website and they localize for me and do Arabic by default, I'm like, wow, this is very foreign. Even when Windows introduced other languages, I was like...

I'm used to using the English app, but that's how I see the default, like everything. I remember once a neighbor came around and asked to do research on something like for a paper. And I think it was like computer viruses. And I think we were like searching for it in the Arabic language, but I don't remember as much, but like I feel,

Yeah, I don't think I have a good answer to you. Okay. Yeah. Well, I just figured I'd ask because we were talking about low language or low resource languages earlier. And, you know, Arabic was certainly lower resource than English because everything's lower resource than English. So I was just curious, like if,

if you used English because you had trouble finding what you needed in Arabic or because it sounds like that was just the way you thought about the internet. The internet is an English phenomenon. Yeah, I did feel that way. Or it was at the time. Okay. Yeah, like de facto, yeah. Yeah. So you definitely strike me as somebody who is intellectually curious and who just...

follows whatever their passion is. You talked about learning how to do cartwheels and stuff like that on the internet. And that the internet represents for you this huge windfall in terms of

Being able to pursue those passions. Can you talk about the education system you went through? My understanding is you were able to go to like these free government provided schools. Yes. Yes. Yeah. The education system is there is the free version for everybody. The government schools that I went K through 12.

Yeah. Even though I missed some kindergarten because I just didn't want to go. But let's say 1 through 12, all free education. And it was in my language. We had English as a subject that we would study. Right now, English is being studied from grade 1. Back in my days, it was being studied later on. And yeah, like...

It was never a financial burden on us to study. Even there are private schools where you can pay a ton of money. They promise you a better future.

I think I'm doing pretty well for myself. So I do think that the government education system is doing a pretty good job with all of those tired, overworked teachers. I salute them. As well as my bachelor's degree, which again, I went to a government university that was also free education. And yeah, my CS degree is...

from there. Yeah. Yeah. And let's talk about getting into CS. So computer science, uh, that is not like a natural undergraduate that everybody gravitates through because it's relatively hard. Like in the U S I think a lot of people go business administration or liberal arts, and you could argue that those are easier than like engineering related degrees because there's a lot, a lot less math. And, uh, the assignments are more like

It's less about like this has to do this or you failed. It's more like, okay, well, they wrote this paper and this paper looks okay. It's not perfect, so I'll give them a B. It's much more subjective, whereas with the engineering disciplines, there are right answers and there are wrong answers often. So I think that does discourage some people. And also CS in the U.S. at least –

They're notorious for having like what are called Weider courses. They're like really hard math courses they put right at the beginning to get people who are not serious to leave the program and transition to like the business school or something like that. Right. What was your experience like getting into computer science and why did you choose it? That's a great question. I don't know if I will surprise you if I give you this piece of information, which is.

When you finish grade 12, your last year of high school, the grade you get decides which university, well, college, which major you get to. Really? Yes. So the most high-achieving students, where do they go? Doctor. Doctor.

Like a physician. Basically, yeah. And then, Mikey, if you're right below that, is that like computer science? Engineering. Science is actually a different college than engineering. So, yeah, like different types of doctors, engineering, and then like law, science, other languages and stuff like that. So there is...

Every year there is the ranking of universities and you get in based on your grades. So my grade was at a point where I had some choices.

And I like personally, because I had some exposure like to just using computers, I had siblings that were working in like I wouldn't call it like programming specific, but like very computer intensive jobs that I remember actually in 12th grade, we had a quiz that said, what job would you like?

have based on your interests. And I remember taking it and it telling me, um, a web designer, because I was like, I don't want to talk to anyone. I like sitting in front of a computer. Yeah. That seems like kind of like an antiquated perception of what a developer role is like, but how wrong was I? Yeah.

But so you chose computer science and then like what was the computer science education like? Yeah. And even you talking about the weeding out courses, things like that. I did not have the freedom to choose my courses. You had a list of courses based on your like your freshman, your sophomore and so on. So we had a list of courses. We had the foundational courses, which is basic.

English language. And those who are not studying computer, they also study a computer subject when they're first year at university. It's kind of an elective. Not really. So everybody has some... Well, it's kind of like...

computer for non-major students exactly yeah but like i said you don't choose what you study um but there are currently there are no electives period basically okay okay and like i'll tell you that does actually make a lot of sense a lot of people are going to be like what no like i want the freedom to choose whatever i want to study and stuff like that but that's more difficult a lot of times professors do kind of know best and they don't necessarily need you to thrash around and

from one major to the other. Like, no, like if you're, if you're serious and you want to like save money and you want to get as many people through that university as possible, it can make sense to have a set menu. You know, just like if you go to like a small restaurant, maybe they only have a few items on the menu. And like, I went to like a Vietnamese noodle place and they only had like three different versions of pho. And I was like, there are only three versions. So like, well, these are the versions that we do really well. And we want to make sure everybody gets something we do well and not some mediocre, you know, imitation of the, the,

the pho street, you know, the pho shop down the street that like has specialized in this type of pho. So yeah, like there is definitely a merit to doing less. If you look at the free cocaine curriculum, it's essentially one curriculum that we want everybody to go through from top to bottom because we know best. We spend a ton of time researching what is important based on the labor market. But anyway, so there are no electives. I will say,

There was one choice where after second year, I had the choice to specialize in

the titles were CS or IT, but like it was like more programming focused or more like, I wouldn't say design exactly, but like just project management slash design slash something like that. They didn't have the like advanced data structures or the advanced programming subjects that we had. Yeah. They had like HCI and yeah. Yeah.

HCI is Human Computer Interface. Yeah. So there was like a bit of choice. Yeah. So that was nice. Okay, cool. So at what point in your computer science education did you hear about Free Code Camp? Ah, amazing question. I will say that I had never coded before I went and started my CS degree, which might be surprising me saying I had access to computers before.

As long as I remember, I think it was mostly like me being someone who just was like going to watch cartoons and stuff without like sitting down with my siblings and saying, what are you doing? How did you do that? Which is something I would have liked to know more about. And it is something that I try to show my siblings. Sorry, not siblings, nieces and nephews. But so me not starting off.

was like something of a, oh, I feel like I'm behind with some of the people in my peers. And I did...

try to like just search around on the internet. Like even back then, Code Academy existed. Yeah, Code Academy. It was like my first- They were like a couple years earlier than we were. Yeah, it was my first like experience of like trying to learn on my own. I will talk about my internship, which is something that I was lucky to do. But then when I, even when I graduated, it was then I was like, I feel like I just don't have the

like the readiness of like, I can start creating a website right now. So it was actually recommended like for me to take the course and I was checking actually my account. I was looking at it and I saw that back in 2000. Yeah. Yeah. Like early 2017 was when I was working and

as a web dev and I was taking Free Code Camp. Yeah. Yeah, that's awesome. So you were able to kind of supplement what you learned in computer science university courses with like more practical kind of like applied web development concepts that you were learning on Free Code Camp.

Basically. Yeah, definitely. Just the fact that I could do everything in the browser, like I could write something, see the outcome of it. It just felt like both something easy to do and also like, oh, I'm doing it practically myself. I'm looking at hints. I'm looking at... So it felt like an all-in-one thing. Yeah.

Yeah, yeah. And that's definitely what we strive to achieve in terms of just having everything right there so you can really focus and rapidly iterate on your code. Not worry about tools. Yeah. Yeah, yeah. Just like remove as much of the noise as possible and just have it be a very focused in editor environment.

learning experience and have it be fully interactive with microsecond feedback in terms of the browser running your JavaScript and immediately running the tests and telling you. But this is not to advertise Free Code Camp. Most of the people listening to this probably already know a lot about Free Code Camp. But I'm thrilled that you found it useful to continue to expand your skills. And you are somebody who...

didn't merely stop at and like, okay, I'm a web developer. This is what I want to do. You've moved on to, you know, machine learning and some like data science, data engineering type work as well. Can you talk about like your career progression since you started working as a developer? Right. Yeah. So I started out, um, like,

doing web actually I have even done Windows application development but my main work was web development and I even threw out like my first job as a developer I even like like

went into Android development, which was like I needed to teach myself Java. It was my first time using the language. I had a base knowledge of C Sharp, actually. And it was something that I could transfer and learn. So when...

And working at a place that had a small group of people, we had the, like, from start to finish of the, like, software development lifecycle of, like, going to meetings with clients, trying to create the schema for the database, creating a data collection tool, a web platform for them, and creating reports, which was, like, something that I saw, like,

like firsthand trying to analyze the data that they collect, trying to create queries and present it in a way that is visually appealing. So I remember like even during my master's analyzing data that's from Stack Overflow and getting their API updated.

something that they have available. I'm sure even now it should be available. So it was something that I was interested in. And me going from my software development to getting the master's degree and thinking, what do I want to do next, was...

I will say, even though I was very interested in, I want to go to the other side, not create data collection tools, but understand the data. I don't think I had the courage to apply for such roles because I was like, I don't have enough information. And my sibling said,

said, here's a job. Please apply. I think you can do well. So just having people around you that can encourage you is actually very useful. And I was like, okay, I applied for that job. I got rejected. I was like, oh, no, I'm leaving. And then he actually gave me another job to apply. I applied to that one. And I saw, I read through who they want, what they want, what tools they want to use. And

After applying, I was like, I need to be not just like I just recognize the tools that they're using. It was not something that I was using exactly. The Microsoft suite of tools of like analysis, even like working with Excel sheets, it was not something familiar to me. Yes, I had some information, but like I feel like I like bureaucratic places are.

I can give you a high percentage of them just using Excel sheets for their data. I mean, that's the case in the US. Like a lot of the world's databases are actually just Excel sheets. That hurts.

As a software person, you're like, I create things to help you organize your data. Please don't do this. So yeah, just like learning about Excel formulas more, learning about Power BI, learning about Tableau, tools that visualize the data, tools that process your data, like

I am in love with the M language, which is the backside of Power BI. I don't know. There are actually two, DAX and then M language. I'm not as friendly with DAX. I like M language more. So just learning about those things using the available courses that are online. Like if this, the place that you are interested or you are applying to uses Microsoft Suite tools, uses Google Suite tools, try and

familiarize yourself because I had an interview, technical interview where I had to demonstrate, um, the, like the tools and the, um, like expertise that I had. Um, even though I was a lot allowed to use the internet, but when you have a limited amount of time, it's better to like say, Oh, I recognize this. I know how to do this. I recognize this, but I need to search exactly the formula and stuff like that. So, um,

Yeah. It's better to review something you have had experience with in real time than to try to learn from whole cloth, a brand new skill. Basically. In a limited time. Yeah. Yeah. So just to recap your advice there, taking the time to familiarize yourself with the tools, like even if you don't get the job,

Spend a little bit of time just getting like going like at least a little bit below the surface on that tool. You don't have to dive super deep, but go a little bit below and kind of see what's down there. And that will at least give you kind of the knowledge of the terrain so that when you need to learn more, you can do so much quicker because you have this kind of like fast pass understanding of the skill.

And then figuring out like, I mean, I always tell people like invest in learning Excel. It's not going away. Spreadsheets are one of the most profound innovations ever in computing. They save so much time.

They were like a 2 or 3x multiplier on a lot of accountants who were spending time plugging data into spreadsheets and then having to update all the cells. Every single time anything changed, they had to recalculate them, which takes a lot of time. And spreadsheets have saved countless human lifetimes in terms of productivity. Probably. So they're not going anywhere fast.

Free Code Camp, of course, has several comprehensive Excel tutorials, Google Sheets tutorials to help people level up. There's a guy whom I've interviewed on the Free Code Camp podcast who runs his own coffee chain and has largely automated it. Okay. He's got all these coffee shop chains. Or he's got this coffee shop chain and he uses Excel and different scripts he's written to basically automate the entire back office type thing. Wow. So he can be super efficient running all these shops. That is amazing. His name is Eamon.

Uh, I can't remember what episode, but I'll definitely look into it. Cause I'm curious. I mean, that's the power of, of spreadsheets. So it is worth investing time in learning those. But, uh, one thing that you said that stuck out at me is that, uh, you were applying to jobs, even though you didn't necessarily feel qualified because your sibling was pushing you, go for it, go for it. And even when you didn't get it, you didn't, you know, whatever.

Oh, I'm so hard. I'm so crestfallen. I can't do this. No, you, you, you stayed emboldened and you leveraged that failure to pivot into another role. Uh, the person thought you might be a good fit for, and you applied for that and you were able to get that role.

Yes. Second one I did. Yeah. That's great. Yeah. So what I want to talk about, I'm very excited about to talk about is a lot of people want to go to grad school and you had the opportunity to go to grad school and you were able to do it while you were working. And this is what I always tell people is like,

Try to get a job after college and then apply. Like they may have a program where they help pay for grad school and then go to like night school, essentially grad school after work. And that is stressful. And that is like, you know, when I went to grad school, like I was fortunate enough to be doing a full time because I was like in a foreign country. I went and studied in China and I just scholarship.

I got a partial scholarship. But I did like Mandarin, intensive Mandarin all morning. And then I went to the courses at night, the MBA courses. So I was able to do that. But most of the people in that, they were there because their employers were paying them. They were running a factory or something like that. And the factory owner was like, yeah, I want you to be as competent as possible at running this factory. So go and attend this MBA course. So yeah, like...

I always encourage people, if you can, to get your employer to help with the cost of grad school. I don't encourage people to just go straight from college into grad school speculatively and rack up more debt, which in Iraq, is it common for people to rack up student debt? Or were you able to get your employer to help you pay?

Yeah, I don't. As far as I know, I do not think we have something called student debt. I mean, you could go into normal debt by studying at a place that's private and needs a lot of money because the grad school that I applied to was not... Back then, the government did not have the program I wanted, so I had to choose a private school that...

When I went there, like asking first time, like, what are the courses? How's the schedule? They were like, oh, by the way, it's after 4.30. I was like, why aren't you advertising this? Like people who are like working like me would love to study in the afternoon instead of like in the morning. So no, I did not get financial like support.

Yeah.

And then I would just leave half a full hour earlier. And even during like exams or like during my thesis, I would take days off or like sometimes do half days because like that was something that.

I discussed and that was something that I could do. Yeah. So that helped me a lot, really. And you got a master's in computer science. Actually, the title is master's in software engineering. Okay, cool. But same, same.

Yeah, software engineering degrees and computer science degrees in the U.S. are pretty similar. There might be a slight difference in emphasis. But the important thing is you got there and you were able to dive a little bit deeper into machine learning, is my understanding. Yeah, so when I first went to like, what is it called? Like an open field.

I don't know, open door thing where you just go and ask questions at the university. I was like, I printed... I wrote a Word document of like, what if we create something like Alexa and Google Home for our language? And then I got some research that's online and I went there and I was like, who's the person who works on human languages with computers, like natural language processing? And they were like...

This person is the main focal point. So without even being accepted at the university, I was like, this is my interest. I want to do something on this. So as a software engineer major, it was not the main thing. But because there was a lecturer who did have the exact...

And I did went into the university's website. I found a few that had machine learning interests and research. And I like even emailed a few of them before again, getting accepted and finishing the whole process. So even those I think did help me. And then talking to that person without even starting classes and then just creating that relationship of like, what should I read online?

Like I have zero information. I didn't hadn't done any research back then. I had not published anything. So like my teacher saying, read this, this and that. So that all of that helped me to just like understand before I like decided, OK, I want to work this on as a thesis, as a end of grad school project.

Yeah, so it sounds like you learned quite a bit while you were there, and you got to... So you were getting practical work experience during the day, and then you were going and learning much more theoretical stuff at night. And I love the idea of taking American products like Google... Google Home, yeah. Yeah, Google Home. Oh, yeah. Yeah, and then just like, hey, why can't we get this working with Arabic? So...

You were doing that. You were working with the professor. You just put yourself out there. It seems like you're pretty fearless in terms of just finding the right faculty member. And they're like, are you even a student? But eventually you became a student, right? Yeah. They were actually really open. They were very friendly. And I am a little sad to say this, that I did not get the same level of openness

openness during my bachelor's, even though the lectures were extremely motivating and lovely people. But they didn't have the same system of like

the student. They were like, okay, this is what we have to teach. Again, they would go out of their way to support us, but I can say at a higher level, management level, they were not given the same level of freedom as private schools were. It just felt like I was being lifted and carried around and like, here, learn this. Here, do this. This is good for you. Probably to do with money. Yeah, that may be like...

For undergrad, I mean, like if you go to any university, like certainly any big public university, you're going to have these giant lecture halls filled with first-year students, second-year students. We had 100 students, yeah. Yeah, and that's just – it comes down to resources. They've got this really great professor who's like won all these prizes, written all this great research, conducted all this great research, and they're like –

That person only has so much time in the day to give lectures and they want to maximize the number of people that get to spend time with them. And it just comes down to, I think fundamentally universities come down to like resources. Um, and that's why you have like in grad school, we were five people in class. Yeah. Yeah. Basically. Much better teacher ratio.

Definitely. Yeah, the level of support did feel different. But like you said, I feel like in a way you are treated like an adult that has an interest. And if the interest aligns with your supervisor, then like...

Even being encouraged to apply to workshops, to apply to conferences, to even have the opportunity to travel while being a grad student, all of that was definitely new things to me. Yeah. Well, I want to dive a little bit deeper into the things you did to build a community of developers around yourself. Mm-hmm.

So you were not living in San Francisco. You were not living in London or New York City or Beijing or one of these kind of like epicenters of technology development, right? Like you were in a place where there were probably lots of smart people, academics. There are smart people everywhere. Yeah.

But you didn't have this critical mass of people. In San Francisco, every cafe you walk into, there's somebody with a laptop out and they've got them open or something, right? In San Jose, every dinner party you go to, everybody's a software engineer, basically. So you're just kind of like...

by developers and it's practically in the air you're breathing. But there are few places where you have that. And I imagine most of the people who are listening to this podcast are in places that are not tech centers. And they do want to be able to find other developers, make these kind of developer relationships and kind of get in

into get a peer group that is continuing to push them and show them, Oh, check out this exciting new thing that, Oh, this is big news because X, Y, Z, you know, that, that like, whether that's like a virtual kind of like group chat type thing or, or a discord server or something like that, or whether that is an actual, like people getting together and hanging out after work type thing. Can you talk about how you built that?

and found other people passionate about, uh,

Computer science, machine learning, things like that? Yeah. I'll give you different levels of answer. For example, I remember back in, I think, 2014 or something like that, I was a bachelor's student. And I remember finding a Google's developer group meetup that I had nothing to do with, like organizing.

And it was someone presenting their master's thesis that had to do with MNIST, which is the data set for numbers. MNIST.

M-NIST, yeah, with an I. NIST. Yeah. It is like hello world for machine learning. Recognizing number data. The hybrid digits thing. Yes. That's it. So it's M-N-I-S-T. I'm going to link to that in the show notes. There's a Wikipedia article about it.

Yeah. So like it was something that I went to. I said hi to the organizer, someone that I follow up until now. Short tangent, I remember talking to the guy who was presenting, saying, I'm going to create Baymax in the future. Because like that was something that I just...

Even back then, without having too much information with machine learning and AI and natural language processing, it was something that I just thought, I was like, this is something that I want to learn more about. So participating in meetups that other people are organizing, that's like the...

best scenario where you're like, oh, I don't need to do that much about it. But I can say that in small places like my city, those dwindle. Like sometimes someone tries their best to do something and then they are no longer in that environment and no one goes and does that. For example, a few people tried to organize Google developer groups

meetups, and I've seen it happen a few months and then quiet, and then someone else restarts it and then quiet. So I guess advice number one would be see if there's a Google Developers group in your general area. So that's one. Another is that...

I found out that there was a computer club, something similar to that, when I was a grad student. So then I also joined that and I was like, I'm going to support the person who's organizing this, someone who's my friend up until now. And that was like, it happened for a few weeks and then it sort of fizzled out. Yeah, that's the thing we usually do. And so...

As you see, like the story that I'm telling right now is that I'm collecting people. So I had a group of people that I knew when I was a bachelor's student, people from my class who are like all my friends on Facebook, you know. So even though there isn't a strong level of friendship there, but these are people that I know that have computer-related interests or just have a connection to computers, okay. And then I try to go to events that other people come

I try to make friends with people who are interested in organizing events. And throughout that, I think it was sometime in 2019 or maybe earlier than that, I remember seeing the Google Hash Code competition and I knew nothing about it. But then I just looked at some of the event videos, looked at some of the questions and I was like, I want to organize this. I went and talked to the

head of department asked if I, we can organize something like this. They were like, okay, we're going to email all of the computer science students. Um, and I said, can we make this public for anyone to join so that I can like create a, I did create a Facebook group. Uh, I shared it with like my group of friends and probably, hopefully they shared it with their group of friends. And then I think we had a turnout of like over 50 people then, um, which was amazing. So, um,

I had the support of people around me to create an event. And I will tell you that when...

me and my group tried to solve the first problem, I got zero, which is in a way I can say impossible to get zero because if you upload the example question, you get some points. I didn't even know that I had to do that. And then after multiple times of like knowing about how the problems are solved, what programming language I'm comfortable with, how at first I try and solve the problem,

I got to a level where I was like, okay, I'm top 10 in my country or I'm top five in my country. So that was lovely. So me trying to organize events and share it with my immediate network and hoping that they share with their network was like level two.

Because I said I'm going to talk about it in levels. And then the final level, which the coding competition did end, I think two years ago. We had a lovely event then as well. But I also have friends who were like,

why don't we have a dev group? And even like right now, I'm in multiple WhatsApp groups of like different dev groups. Some of them like people from abroad, some of them people from here who organize events, whether it's like once a month, like a few times a month, they just like bring people in the community to talk.

and there are snacks there. You just get to know people. And it's a lovely thing that they support the whole... For me as a woman in tech, I am interested in seeing other women in tech being brought up and promoted. So it's really lovely to see that it's something that they also care about. Yeah. Yeah.

Wow. Okay, so you used the three levels. I love that line you said, I'm collecting people. Essentially, you're meeting other people in the field and you're becoming kind of like a known quantity to them and you're learning about who they are and you're using that. You can use that to branch out and like...

Just knowing a few people, you go to an event, you can meet more people, and you can just very rapidly expand your network. I mean, we literally call it networking, but it is kind of like computer networking. Oh, definitely, yeah. And if you support other people in their endeavors, whether it is organizing a meetup, like even going on meetup.com, which I think is a website that people do put their...

events on. But even if you don't, if I go to the advice territory of like, okay, I find I found nothing, I don't have a WhatsApp group, I don't see other people organizing, my university is not supporting me organizing something, I would say whichever platform you find more comfortable, whether it's Facebook, something that

you and your friends use, whether it is something more like chat, whether it's like Messenger, Telegram, whatever it is that you and your group of, I'm guessing, students or just people that you study with, even if it's like you go on Free Code Camp's forums and you say, how many people are in this country? Or, don't have to be in the same country, you find a Discord and you just create a community of people.

people online that you find people who are interested in learning this language, learning this tool and you just say, okay, let's have bi-weekly meetings. Even if we cannot make it at the same time, let's just have a chill chat session, a chill work session. Even for me, doing live streams is my way of creating a community of people who are interested in the things that I'm interested in or even people who want to

Like learn and find people who want to learn at the same time, even if it's not the exact same thing. So you can see like no matter how much of a programmer you are, we're still social creatures. And we do crave having this community of people to support and have them support us. Yeah. Yeah. And I think what you said there about just learning.

putting yourself out there through these, like your live coding, right? You're going on Twitch and you're coding live and you're talking with chat, you're interacting and you're trying to do coding challenges. I live fail, yes, a lot. Live fail. Yeah.

And I did Twitch for a while and a lot of the early FreeCodeCamp development was done on Twitch with me like, oh, okay, now I'm struggling through this. And it was kind of humiliating. It's a humbling experience. Yeah. Getting stuck on something really basic and it's always great when somebody from the chat is like, oh, you forgot this or like, oh, did you look into this library? Yeah. Wow, that's a huge help. Definitely. Yeah.

Yeah. Maybe you've got like two or three people watching, but it's still, it's, it's nice. It's accountability, right? You're, you're not going to go in like, I'll tap over to Reddit and start reading like, you know, news or show on video games or something. When you've got a bunch of people watching you, you have to perform. So it's a way of kind of putting pressure on yourself to continue to learn and do your best. Right.

It is my, it is my, um, like, I feel like I'm cheating my way of the accountability part. Cause I have this schedule of like Friday and Saturday, I'm learning this. Um, and if I don't go, um, it's a way of like, oh, I broke a promise. Even if one person was waiting, expecting, um, and, uh, like it's been, I can say two years of like,

frequent and regular streaming at those times where I'm like, it's a part of the routine now and I expect it from myself and I am actually learning even though I am...

I'm going to put quotes in that as well. A lazy learner of like, I don't always do my best. I don't always give it 100% of focus. But when I'm like learning on a regular basis and I, okay, learned about AI agents in this blog post. And then a few blog posts later, I learned another definition or a different use or a different library. And it just like,

cements itself in your brain and like, Oh no, I'm very familiar with this back then. I just didn't even know how this was, what it even meant. But now I'm far more familiar. I can think about ways of using it. Um,

Yeah. So just like, I feel like I'm cheating my own system, like finding a process of what works for you. And because as developers, as people working with data, with software, all of that, the world is developing and like evolving so fast. And it feels like

we're running to catch up. Um, so I feel like the learning, if you want to continue in this, um, area, it's something that we cannot like just sit still. Um, so just even having like, um, blog posts on food, Coke, Coke camp, um, talking about more recent things when it's not as easy to integrate it into a full course, all of that is, um,

Yeah, it is lovely to be able to have a refresher when maybe you don't get the chance to take courses or learn new things during your job or even at school. Yeah.

So it sounds like to an extent you're using streaming as not only an accountability mechanism for yourself. And I should point out there are very few big prominent streamers that are streaming themselves coding. It is a more niche thing. There are, but it is niche. There are a few. There are a few. And a lot of them are just talking about...

programming or maybe reading articles and reacting to them and things like that. But then there are the people that are actually like, okay, day 28 of me building this new operating system. Yes, there are some lovely people creating tools. I'm more of a – I like to study and learn together and try to test things. But there are people creating actual products on –

Twitch is more of a place that you can like search by software development. YouTube, I have a harder time finding people. Maybe it is my feed, but there are some very, very lovely VTubers that are actually like teaching STEM on YouTube. Yeah. And when you say VTuber, you're talking about like people that are like virtual tuber, like they're using like an avatar. Yeah.

Yes. And there are humans behind them. I've been asked, are you AI in chat? And I'm like, oh, thank you for thinking I'm AI. Yeah. Yeah. There are people who have like PhDs behind a VTuber reading a research paper on stream. Like there are so many amazing people that I have found on both YouTube and even to Twitch. There is a very lovely community that works.

is like into learning, whether it's like astrophysics, like Z, someone that I follow and love. Yeah. Like V theorem is a group of like very smart people that I just admire and look like from behind my computer and learn from. Yeah. Yeah. And I think like there are probably people out there who have like a parasocial relationship with you and that they watch your stream and stuff like that. And then they probably have,

uh, other people that they have a similar relationship with. And then there are people like me. I have a parasocial relationship with a lot of people. I watch a ton of programming related topics on, on YouTube and I listened to a lot of programming podcasts. So I kind of am like a fan boy of a whole lot of other creators. So it's cool to like, come on on. Uh, and one of the things you do when you put yourself out there and you create, you know, for lack of a better word content, I don't like that word because it's like, it's filling some, some,

vacuum or something like that but i is it not art i don't know the art of the interview the art of uh you know going digesting an academic paper and synthesizing it into something a relative lay audience can appreciate and understand you know the art of talking while you're coding and trying to explain why you're using a specific data structure or a specific library like uh but

But yes, I'm not going to try to exalt this podcast to like being high art or whatever. It's not like Picasso standing there like, you know. Did you know they created a Grammy category for podcasts? Ah. There we go. That's art. I guess so. Yeah. I'm going to nominate this. Well.

Well, I do know that we have a three-time Grammy-nominated saxophone player and arranger who just published a course on C++ on Free Code Camp. How chill is that? Yeah. Oh, wow. Building like an audio plug-in in C++. So that is the caliber of the folks we have up in the Free Code Camp community. Wow. Three-time Grammy nominee. Yeah, super chill. Yeah. With a community of millions of people, I'm sure that, like,

there isn't a place you have not reached on this globe. Yeah, or...

Yeah, we definitely have parasocial relationships with you. But yeah, we're going to be chill here. So this is a practical tip for people listening at home. And from here, like we've already talked quite a bit. I want to dive into the rapid fire practical tip. I always like to end with lots of practical tips for people because that's what they tune in for. They love learning about people, but they really want to like, what are the takeaways? How can I apply this to my life? So one thing, just...

Getting on Twitch, of course, I talk to a lot of people like just stream on Twitch. It's not that hard. Like you don't even necessarily have to put your face on there. You can put a VTuber type avatar if you get the software and download it. HoloLive, I think, is one of the ones that Naomi on our team, she uses that. Wow. Actually, I'm not sure she uses that. But you have to learn how to use like OBS. What other tools do you use?

Yeah, great question. So yes, OBS and I have learned from YouTube. But I am using a model that I actually, I can say I do not own. I did try to create a VTuber model using the program that people use for modeling.

So it's something that you can find if you search VTuber model. I did try doing that, but I found an available application called Zepeto, something like that, Z-E-P-E-T-O, where you can design your character. I just love the fact that they had like Hidzabi options.

So it does the face tracking. There is no hand tracking, which is something that a website like Kaleidoscope could do, something that I also tested. I have tested probably five, six, seven tools before I landed on this version where I'm sharing the screen of my phone with OBS and I'm making the background green screen and then just putting it

on a background that I created on Canva. So I did not, I can happily say I did not pay for anything, even though there are people. Yeah, there are. I did pay for my microphone, which is a very fancy one that I got almost like a month ago. But I used to have another one that I got on Amazon for like less, definitely less than $100. It was a very functioning microphone that I used and

I can say over 10 years. Yeah. If anybody's looking for microphones, if anybody's curious, this is the SM58. This is a classic microphone design from like 60, 70 years ago. I think it's like 60 bucks, maybe 80 bucks. Great. But like...

This is what like music, like Paul McCartney, Rob Halford, Judas Priest, like a lot of people use this mic for singing. And I find it works really good if you have like a fixed position. Is there a blog on Free Code Camp about like the tools that you use for podcasting? Maybe. I can't remember if I've written such a thing.

Yeah, because it's actually very interesting that people may want to try and dip their feet in. Because even I have thought about, should I podcast? Because it is an interesting new thing through COVID times. But yeah, so I was able to test multiple tools. I used to use Android Studio as a virtual phone.

on my computer, but then the app stopped working on there. Now I'm using... Because I'm using a map, I'm using something called Screen Copy, even though for Windows there are so many tools that help you show your phone on your computer. But I had a problem with connecting Android to Mac, so I found the solution there. And yeah, so I'm lucky to... Now I have a setup. So like...

When I do my morning prayer and I come to stream at like 5 a.m., everything is ready and I just click stream because that's my normal time of streaming. If you would like to write an article about your setup, we'd be thrilled to publish it for sure. Oh, that would be lovely. So definitely consider that. And if you do that, I will come and add that to the show notes of this video.

Thank you. And of the audio edition podcast where most people listen to it. So some questions I have for you. You have continued to expand your skills. You are a big advocate of lifelong learning. Give us kind of your learning stack or like how, what is your information diet like? How do you stay on top of the rapidly changing space that is like machine learning and AI tools?

For me, that's a difficult question because for me, I try to have, like you said, a diet of I'm interested in academic articles that are relevant, but it has to be something like the foam at the top because it's not really easy to read every article that comes out. So like reading, having a deep dive in,

deep dive into like the deep seek, um, paper or the whisper paper. Um, so these are things that I try and, um, learn, but for me to, um, for example, uh, something is, I would say as recent as, uh, agentic AI. Um, I found that, uh, hugging face has created a course for it. And I, uh,

started doing that. And one advice that I give to people who like myself go into like course learning hell is that

try and not give up too easily. Like, let's say you start learning, you go 20% and you're like, this isn't exactly what I want, but I would say try and like power through and at least reach 50% before you decide on, okay, I'm going to hop into another one because that is something that I can say I have done. So, yeah,

Places like, I'm sure if you go to Blue Sky, you can find a group of like, I'm interested in learning about the research that NASA is doing on Mars. So there's definitely a group on that. So trying to curate your feed to give you that information, maybe like X might not be the best place, even though it used to be. And even like having trusted websites. For me, like when I'm interested in...

models, I would go to Hugging Face when I'm interested in more software-related things. I might go to the blogs on Free Code Camp. So having trusted websites that you could go back to, even having people you follow on YouTube who give you weekly updates on tools that are available. Even when I learned about

The fact that Amazon allows you to take one free course on their website. I was like, I'm going to make a short about this because people need to hear about this. So I try to share information that I get, but it's really not easy to digest everything that's out there. So try and be picky, but also...

I sometimes I just want to watch a K-drama and not learn new stuff. So I'm not going to just be like, say, learn all the time, 24 hours, read and study. Yeah. Yeah.

K-drama. So my mom loves watching Korean dramas. Yeah. And Korea is such a fascinating culture. Absolutely. You've been studying Korean for a while. Not to totally go in that direction, but you did mention K-drama, so I wanted to dive into your language learning. For people out there who want to learn a foreign language, let's say they really like K-drama and they want to learn Korean. What should they do? How do you go about it? Mm-hmm.

Good question, because if you can believe it, I started learning when I was in high school. Okay. Yeah, so for me, it's probably been like 13, 14, maybe something like that. I started learning Hangul, which is something, the Korean alphabet, which if you like Google it now, you can learn it, I will promise you, in 15 minutes. It's a...

alphabet that's created for the people who had a hard time learning the Chinese alphabet because it is very difficult. So King Sejong decided to create something that normal people can learn. So yeah, that's like the history. So I had a book and I started memorizing the words on there, probably gave up early. And for me, the way I learned the English language was through just...

I hope that I correct. Kind of like the same way an LLM learns. It's just like huge corpuses of English media, essentially, like all those Disney movies, all those Shrek movies. So now you're doing the same thing with K-drama. I am. The type of entertainment that I like, I try to get in Korean. For example, someone was telling me, did you see this series? I was like, yes.

I feel like it's a waste if I watch American TV shows because sadly I can watch a summary on YouTube. But like I want to listen and watch K-dramas, Korean movies, which are just amazing. You can see Parasite, for example. So yes, I have like, again, curated my entertainment to be more Korean, but my news is mostly English sources. So yeah. Yeah.

For you, Korean is like the fun language and English is the serious work language. Yes, yes, yes. That's cool. Yeah. I think it's really cool that you're continuing to learn a lot of different things and you're not resting on your laurels. One thing you've said that I thought was really interesting is...

It's important not to tie yourself worth to your job. It's okay to have great days and don't expect every day to be a happy day. You can, you can kind of strive to be content and you can optimize for contentment or you can optimize for kind of the more stressful, regularly updating your skills, giving you a competitive advantage over others. The, the comfort versus challenge. Can you talk about that a little bit? Like,

At what points in your life should you be seeking comfort? And at what points in your life should you be practically seeking challenge? Should you flip-flop between the two? Yeah, that's a really great question. Because I...

Let's say you are someone who's interested in changing jobs and you're like, when I become a programmer, all of my days will be high days. Everything will be the best. And at a point, it might be, but the hedonic treadmill exists. Hedonic treadmill, yes. Yeah. Very important concept everybody needs to learn about. Basically, your body, your mind adjusts.

Like you're used to having nice morning coffee every day. And then like when your coffee's not there, you're like, what the heck? But if you never got adjusted to like always having nice morning coffee and you just drank water, you wouldn't be all upset that you suddenly didn't have coffee because you wouldn't realize what you were missing. So like humans always want more. Yes. And so like allowing yourself to not always be like, this is my best day. I'm going to do a hundred percent of my like energy.

my efforts into the job that I'm doing. Um, I think trying to be kind to yourself, uh, I have a tiny advice of like, it's been a few months, maybe more than six that I'm, I'm doing a work diary, like starting my day. I do 10, 15 minutes of like writing, um,

Like this is what I did yesterday. This is what I'm gonna do today You're gonna do great. Good luck and try to be kind first like this Talking planning but like wording it out I Need to work on this someone is waiting on me on this stuff like that and then trying to write a to-do list So not expecting yourself to be at a hundred percent allowing yourself to have those 50% days like maybe you

you have like a energy of like a monthly cycle of different types of energy and that's normal. So try and... Here's a cheat. Try and give yourself... Give your 70% of efforts to people who expect things from you and try your best to be nice and people will forgive the rest of the percentage. And on days when you feel 100%, those people like...

would be surprised. But even then, try and not give 100% of yourself at anything. Ah, interesting. So always try to give 70%. That way you have like kind of a 30% threshold. If you're only feeling 70%, it still seems like you're giving 100%. Even if you're feeling...

50%, they're like, oh, the difference is not that much. But from 100% to 50%, there's a lot. But to even answer your other part of question about comfort and challenge, for me to have worked at a place for more than four to almost coming five years, there is a level of comfort at your job where you're like, I know everything. I know what comes during this time.

like quarter one, quarter two, quarter three. Um, so yeah, yeah, there is a cycle cyclical feeling of like the type of work. So you're like sitting in comfort and being like, I'm safe here. Uh, it's not always the case. Like if you look at the global, um, environment, uh,

whether it's academia, whether it is organizations, you can see that some of the funding is going dry. And it is something that I have seen people being affected and laid off. And like being someone who has this, I will say 10% of your week being, I'm learning something that will improve my career. Even if you're not,

if you are someone who's like, I'm happy with my job, I'm comfortable with my job, but I'm still learning in case, then you don't feel like, okay, I'm a risk taker. I'm trying to go for something higher. And then when the

like the thing happens where, Oh, my job is no longer there. Um, then you are much ahead than some people who just actually sat in that comfort. But if you are someone who's like, um, you know what? I know if I change my job, I'm going to get a higher salary. And, uh,

The risk for me, like I'm, again, probably someone who tries to be safe of like, okay, I have a job, but I'm going to apply to many things and see which one I get back to me. So if you don't have that level of privilege, then yeah, I may not have an exact answer. Yeah. Yeah. No, that's,

Exactly what I was hoping to hear in the sense of like it's something new that I hadn't really thought about. But to some extent, by having options, by cultivating new skills and cultivating those options, you're actually a lot more comfortable. There are lots of people that probably think they're comfortable that are kind of just like chilling. Yeah.

wherever they are, just relaxing and enjoying the fact that they have achieved a certain degree of mastery over their job and just feel comfortable doing that. But are they really comfortable doing

When they have that kind of sort of, uh, Democles hanging over their head, they can snap at any moment and have to go back on, on the job market and they may not be prepared. Yeah. The rug was pulled. Yeah. Yeah. The real people who I suspect are content are those who do spend like 10% of their week continuing to invest in their skills and, uh,

plan contingencies in case the rug is pulled, in case funding goes away, in case there's a giant scandal revealed and the entire organization goes under all of a sudden. That's possible. That happens a lot here in the States. There used to be safe jobs, but honestly, in this economy, it's almost like gone.

Yeah. Like I was listening to your podcast about a person who was working at Microsoft and talking about management changing and organizational structures changing and just people being laid off. And like for me, Microsoft seems like one of the safest places out there. I mean, it's the biggest corporation on earth. That's not safe where it is. Yeah. Yeah.

Yeah. So, uh, so, so that, that is definitely something I'm taking away from this conversation is it is absolutely worth continuing to invest in yourself. Uh, it is worth putting yourself out there and what you've done, where you have been able to preserve your relative anonymity. You're like a pseudonymous, pseudo anonymous, like a person on the internet and people can tune in. They can cheer you on. They can help you. They can learn from you. They can develop this relationship with you. But at the same time, you, you,

still get to be a private individual. That is great. I wish I could do that. But as the executive director of a public charity, I have to actually... You could not be a VTuber? Yeah, I mean, I guess I could be a VTuber. We'll look into that. In addition to doing this. But yeah, I have to be a public person. So I don't have that luxury. But it's not like I'm complaining. But what you said about, I think...

building community. You said you're collecting people. You talked about trying to get events going. Attending other people's events is great, but why don't we go out and build our own events? Why don't we get our own WhatsApp group going or our own Discord going about a particular topic and kind of just grab communities? And then people can be parts of so many different communities. Like, I'm

I'm a part of, I don't know, like 30 different discords. Right. And usually I'm just lurking in there, like reading, reading what people are saying and stuff and occasionally interacting. But like, it's a great place to like get updated information based on what you're interested in. Yeah. And it's a great place to form real authentic relationships with people who may not be in the same city as you. I love going to conferences. Yeah.

I am going to go to so many conferences this year. I'm giving toxic conferences. Yeah. And, and, you know, I, I encourage everybody if they have the resources, if there are conferences nearby and if there aren't conferences nearby, see if you can, this is very important. Don't break the bank. Don't spend a whole bunch of money trying to put on a conference. Don't take like financial risk, but see if you can,

find somebody that might be able to like sponsor the event and some, just some office where you can meet after hours, uh, and have like a meetup group and just, yeah. Like, like, uh, so what you're doing, I think is awesome. And I want more people to do what you're doing. And that's why I was very excited to bring you on the podcast, CL code life.

Yeah, it is truly like an honor to just have the opportunity to talk to you. And thank you so much for like allowing me to talk to you and to share whatever I have with your audience. Yeah, well, we're proud to continue, you know, have you be a part of the Free Code Camp community. And I feel personally very proud that my, the learning resources that I've been a part of helping make possible through the Free Code Camp community, through our

thousands of open source contributors and authors, course creators, all these people that we have helped supplement the education that you've gotten through the government university that you attended and the graduate program. And I hope that you'll continue to learn through Free Code Camp as well as you continue to expand your skills. If I can compare Free Code Camp to something that anyone can recognize, it'd be like you guys are the Mr. Beans of TV.

Like if you go anywhere, someone will recognize it. Even because like I have people around me who don't even have a CS degree who are like, oh, I'm interested in learning what coding even is. And like, I know the place for you to go. So even like with the family around me and even with online people who are like,

okay, I'm just like interested and see if I'm curious if I want to learn more about it. Like here's the place for you. Yeah, absolutely. So thank you so much for, for allowing free, like helping free resources be available online for people like us and many others. Yeah.

Thank you so much for your kind words, Codelife. And everybody, please check out the links in the video description. I've put links to some of the things we've talked about in there, including one of Codelife's videos. And I hope until next time, until next week, you all have a fantastic time and happy coding. Happy coding.