I get this, like the top requested type of video that I d'oh.

Uh, and the funny part is I get more requests to do Q and A videos.

Then I get actual questions that I would answer in a Q and A so I can't really do Q and A videos without the queue.

So if you've got some questions, feel free to leave them below.

Otherwise, I'm gonna answer some of the questions that I have from a community question that I asked, basically just YouTube dot com slice and X large community.

There's a place there I can just post stuff.

So some people have asked some questions.

There.

I've got further ones, or if you ask the question there and I'm not getting to it here, it's just because I can't answer every single question in one video and try to use do some that were somewhat related to each other.

So anyway, let's jump into it.

So the first question here is what is the most challenging problem you've ever come across and coding, And how do you solve the seemingly unsolvable?

I'm going to assume that's not assertion that I actually solved the unsolvable.

I'm going to assume you being just how do I solve hard problems?

Because I actually don't really solve that hard problems typically.

So, um, private most challenging problem in programming is not necessarily like any programming concept or something to learn.

I think the most challenging thing is is like technical debt, but not just technical debt like it's.

It's sort of avoiding technical debt, but it's his balance.

You always have to strike between writing the most efficient code possible, but also just get the thing done like just actually make the thing.

Like a lot of people, I think, focus too much on trying to write something to be super efficient.

But yet it's not readable, right?

So writing efficient code, actually getting it done and then, like, maybe it's a triangle.

We'll see if I think of any other answer here.

But, uh, so getting it done, writing the writing, good, efficient, nice quality code, but then also writing code that can be built upon moving forward and not just moving forward today and tomorrow and the next day.

But like any year from Now, when you go back to revisit your code, are you gonna understand these lines that you wrote?

So a lot of times, like on Stack overflow, like certain question answers that get posted there, the code is really cool.

It's super efficient, runs super fast.

But if you didn't know that, that code was an answer to the question above, you would at first glance be like What the heck is that like?

It's sick from the nastiest code ever, but it solves the problem really quick.

And that seems to be what most people go for is like, How can we solve this problem?

Is few lines as possible?

And that's not always the answer.

So, anyway, is getting back to the question.

Honestly, I think the biggest problem is just projects growing over times.

It's really easy to come up with a really novel and fancy answer really quick.

It's also really easy to write really bad code that is not going to be modular, and you're gonna have a really hard time building on it, moving forward and right completely alleged rubble code, that sort of thing.

I think that's the biggest problem.

I think it's just some sort of project that you're working on, growing it over time and and having that kind of foresight as you start working on that sort of a problem because a lot of times you just at least for me everyone's a little different.

I definitely think there's, I don't know who at first came up with the there's makers.

And then there's menders concept with programming, but I definitely think it's true.

I'm sure each of those has their own little category, and I'm sure some people are going to assert that they're both things or whatever.

But I definitely tend to see there are two different types of programmers, people who just make stuff really quickly and are really good at, like prototyping and just getting this sort of N V P type thing out.

And then there's the menders who come in and can actually keep a project going long term.

I'd like to think of myself as more of a maker, so maybe that's why I think that's the biggest problem is because I'm not good at it.

I just I'm really good at, just like bopping out a really quick project, solving a problem But then, over time, I'm not very good at keeping a project growing.

So anyways, think, answer that question next question here from Max.

What programming project years that you're most proud of.

How did us a few questions?

Apparently, uh, how'd you get into ml and what's your favorite type of program?

I'm not really sure what you mean by type of programming to do there, but probably the most.

I mean, obviously, I think I'd have to go with the g t a five.

Siri's as faras like, what is probably one of the larger projects that I worked on that I was, you know, kind of pouring into the previous question.

Um, that was a project that I had at least enough foresight.

I mean, it's definitely a mess in the current state tickets in, but that was at least a project that I knew kind of like, Okay, I need this to be its own separate little module and this and this.

And so I did pretty good at actually making it happen, but also continuing to work on that.

That was probably other than like python programming that net like Syntex dot com or something.

that project, though, in probably more so than Python Parker and that I put mawr continuous hours.

Like more time spent continually, like building on top of it over time than I have on any other projects.

So I guess I'm the most part of that for the fact that I was able to do that and not just like I quit.

So, um, anyway, so, yeah, probably that How'd I get into ml?

It just kind of just sort of, like, eased into it.

Honestly, Like I got in the ml before there was like, I mean, deep learning was a thing.

And, like, neural networks were a thing.

But they weren't big, right?

They didn't really start getting big until, I don't know, maybe, like 2015 and then onward.

I mean, I'm sure they were They existed like something when I was in a neural networks in 2012.

I'm sure you were.

That's great.

That's cool.

But really, they didn't really get big until a little bit after 20 some time, like 2015 on, I think.

Anyways, um, so really, I was just interrupt.

I just found it just awesome.

Like with psychic.

Learned the things that you could do with just really basic ml algorithms.

I think it's unfortunate most people are many people don't really care to learn those anymore because we have deep learning and everyone thinks deep learning solves everything.

But if you've if you've ever done anything with, like, psych, it learn and just like a nearest neighbors K nears neighbors is the best algorithm ever.

And it's so basic, but it can It can be, you know, parallel process to, you know, however many processors you haven't a consult awesome problems, similar problems to what you would do with neural networks and stuff.

So anyways, not what you were asking about.

But yeah, I just I found it really fascinating.

I just think like she's, like, really pretty darn simple algorithms that can do really cool things.

Um, yeah, and one of things I really like to get into that.

I've just never gotten around to his Grady in boosting, like with x g boost.

Um, I really need to spend some time there as well.

Um, it's just yes, something about neural networks that just draws people, and I think, but Yeah, I just got into it.

Uh, really with sentiment analysis.

That was my first real ML type task and like doing things like with finance stuff like that.

That's how I got into it now.

Favorite type of program.

I just don't know what kind of question that is.

So I don't think I can answer that.

I don't know if you mean like, competitive versus I don't know, building things I don't know.

I just like to build things.

I don't think I have a good answer for that.

Moving on.

Uh, how do you get around Bad documentation or bad?

Like Repo?

Um, maybe that's just that's a huge question.

It really depends.

It depends like do I have to use that library?

If I don't have to use that library, I will go use a different library.

If I do need to use that library for whatever reason, my next step would be to find, like samples of code from other people.

So see who has used this project before.

And honestly, I usually do that before I'll even read documentation in the repo, to be honest, like I'll just see.

Is there some somebody actually used this code because that's where I like the documentation is one thing, but all you're you're getting like, these little slices.

Whereas if I can see how someone has used it in, like an actual project, um, it just clicks better for me anyways.

But some documentation is exquisite, like the pandas.

Documentation is is really good because that's what you're you're just using me like really unique tools on a date, a frame.

So like in some cases, documentation makes a lot of sense, but in many cases, it doesn't quite make a lot of sense.

And then also, if a thing like generally the documentation becomes useful to me after I've learned how to use the module with library wherever we want to call it, so then I'll go to documentation.

But first I have to learn, like the basics of using this thing.

And usually I go find sample code first anyways.

But if for whatever reason the documentations bad more samples or read the source code, you congee just go into the source code and read it.

The other thing I tend to do is just to see what are my options.

You can use the dir d I R.

And then in parentheses, past the thing that you're curious, what are all the methods that we can do on this thing?

I'll just do that and see.

OK, Aiken.

Usually they're properly named, so or you can guess what it means.

And then you can just kind of trial and error just kind of play around and, um, and just figure it out that way, honestly.

So, um, hopefully that answers your question.

So next onto the last question here, what approaches do you take in order to get a better understanding of new programming concepts that might be confusing and frustrating?

The best tip I can give their eyes.

Take your time.

I think people get, like, way too enamored in like just sitting down and just grinding a thing out when sometimes the best thing you can do just, like, walk away, go to sleep, take a shower, eat, eat some food.

I could definitely say like like almost daily, I'll be like working on a problem, and I really want to finish this thing.

But I'm getting hungry and I just I'm like, I just wanna really get it done and I tried to grit my teeth and then eventually it's like now you need to just just go eat, Just just walk away.

Um, so sometimes the best thing you can do is just walk away.

And I'm like, sometimes I'm you know, if you're just hungry, at least for me, my brain just starts getting worthless.

We're tired, so I just kind of walk away.

But even on, like, really tough, challenging concepts and stuff, the first thing you need to take into consideration is maybe today you're not going to get the whole thing.

And the best thing you could do is just kind of walk away.

So just keep tight taking, like little doses.

Also, Um, generally, if a thing is confusing or frustrating, it is confusing or frustrating because you have attempted to skip a step.

So I've never come across anything, be it programming or any other concept in life.

Where if if I did things in the proper order, it it was still confusing.

Yeah, so So, generally, any time anything's ever been confusing to me, it's because I'm trying to understand it concept before I've understood like a pre requisite concept to that concept.

So a lot of things like math related things.

Programming related to, um, one of these things.

If you find yourself confused, then you're probably you've skipped something, and then the next thing you want to do is figure out what was that thing that you've skipped because you've You've probably skipped a thing.

Um, other than that, if that's not the case again, just go to sleep or, you know, take the day off or whatever and try to approach that problem again.

But, yeah, I think a lot of times people just want to grind things out.

I mean, myself included, I'm not pointing the finger.

At least that's That's been the truth for me.

And most of my largest gains have occurred while I slept, and I just revisit this thing the next day.

Okay, so, uh, that's a few questions answered.

I know there was a lot of other questions that people had, um, posted already.

If you've got your own question, fill for you, leave it below.

Um, I'll get to the other ones.

I just felt like these were all kind of like coding related questions.

If you posted a coating related question, I didn't answer it again.

I might still answer it.

I just didn't want to answer like, 30 questions.

All in one video.

So, anyways, finally, a quick shout out to recent Channel members Uh, Ridean Ferdinand Mohammed Al More backwards are Michelle Berardi, Matthew Bloem, Ivan Buck, Nick Jean ends or fidget, Jackson Carril and Henry Demeanor.

Thank you all very much for your support.

You guys are amazing.

Um, I think I said new members.

Actually, I think there was only 1 to 2 people that are new members.

The rest of these are re re upping members, so, yeah, thank you guys very much for your support.

Um, till next time.