Demystifying AI with Louis Bouchard on The HackerNoon Podcast

Written by podcast | Published 2022/01/05
Tech Story Tags: podcast | ai | hackernoon-podcast | artificial-intelligence | future-of-ai | hackernoon-top-story | podcasting | youtube

TLDRThis week's episode is sponsored by Bybit - one of the fastest-growing cryptocurrency exchanges, with more than 2 million registered users. Vote for Louis for the Hacker Noon Noonies 2021 Awards he's nominated for: computer-vision, data-science, deep-learning, facebook, iPhone, and the iPhone. Read the NoONIES: NOONIES.COM/tagGED/AI - WATCH the latest episode of The HackerNoon.com/taggued/AI.via the TL;DR App

Louis Bouchard is a master's student by day and an Artificial Intelligence content creator by night. He is the creator of the YouTube channel, “What’s AI?”, which has 23k+ followers today. Amy chats with Louis about his content creation journey, what motivates him, and how he balances student life with YouTube video creation.
On this episode of The HackerNoon Podcast:
  • How long has Louis been an AI YouTuber? (01:46)
  • What has been Louis’ favorite use case of AI? (06:43)
  • How did Louis grow his AI discord channel? (11:10)
  • What is Louis’ YouTube strategy? (15:11)
  • How does Louis structure his day? (21:10)
  • What is the future of AI? (27:00)
✨ This week's episode is sponsored by Bybit - one of the fastest-growing cryptocurrency exchanges, with more than 2 million registered users. Learn more at bybit.com
Vote for Louis for the Hacker Noon Noonies 2021 Awards he's nominated for:
Find Louis online:
READ AI STORIES: HACKERNOON.COM/TAGGED/AI 🏅
WRITE YOUR FIRST STORY: HACKERNOON.COM/SIGNUP 🏆
CHECK OUT THE NOONIES: NOONIES.TECH 💚

Podcast Transcript (Machine Generated)

[00:00:00] Amy: Hey hackers. This hacker noon episode is sponsored by Bybit the cryptocurrency trading platform that takes buying, selling, trading, and earning crypto to the next level. So visit bybit.com to learn more. And, I want to let you know that the hacker noon shop is now and officially open. So go over to shop.hackernoon.com for all of your hacker noon merch.
We've got t-shirts. We've got joggers, we've got notebooks. We've got laptop cases. We have stickers. We have it all, whatever your hacker, heart desires. So go over to shop dot hacker noon.com to pick up your hacker noon merged today. And don't forget to vote for your favorite startup at startup.hackernoon.com.
And for your favorite new news nominee at noonies.tech onto the episode.
Ooh, what is AI? Hello hackers. Welcome to another episode of the hacker noon podcast. I am here with a very special guest Louis Bouchard, who is a top contributor at hacker noon and has been nominated for literally eight new knees awards.
Let me read them out to you. I have it like literally written down, cause there's so many. I can't even remember off the top of my head. All contributor of the year in the categories of computer vision, data science, deep learning, Facebook game development, Google, iPhone, and machine learning.
Welcome to the show, Louis, how are you doing? 
[00:01:36] Louis: Thank you. I'm great than you. 
[00:01:39] Amy: I am doing fabulous. So I have a few questions for you. Number one, you are an AI YouTuber and you have a channel that is called what's AI. So how long have you been doing that? 
[00:01:53] Louis: It's it will be two years in January, so yeah, two years, but I started using.
A, an artificial voice. So like the real channel actually began in May 2020. So a bit less than two years. Yeah. 
[00:02:10] Amy: Oh, okay. You started, so you started not even using your own voice, you programmed an AI, or are you using an AI to do the videos? 
[00:02:19] Louis: Yeah, I use the text to speech just because I loved the idea.
Well, at first I wanted to teach AI only like the basic concepts of AI. What is machine learning? What is deep learning? And I liked the concept of explaining AI using AI, but yeah, it didn't sound human yet. And at the time, at least now it's already much better. And also I want it to stay anonymous.
Just, I actually do not know why at the time. I changed that a year ago, but yeah, I started anonymously. So that's the reason 
[00:02:55] Amy: was your reason for starting the channel? 
[00:02:58] Louis: The, well, there are a few reasons. The first one was to to make sure I learned the country. To make sure I deeply understand the concepts of artificial intelligence.
I actually started sharing and explaining what is machine learning while I was learning about it, just to. To deepen my understanding and be sure I could, if I could explain it, that will mean I will get it. So yeah, that's what that was the first reason. And the other reason was to why that there are two other reasons.
The second one was to improve my English speaking because I'm a native French speaker and I wanted to become a bilingual or, well, at least. And the last reason was to, it was because I was about to start a master's degree in artificial intelligence. And I knew that reading research papers was important, but I never actually read any of them.
And it was scary. Yeah, I started reading them and it was hard to motivate myself to read them. I figured why not try to explain them to be, sir? I read them completely every week and that I understood what was in the paper. 
[00:04:13] Amy: Right. Okay. That makes sense. Wow. The master's in AI. I didn't even know that was a thing.
Of course, I guess it is a thing, but I never really thought about. 
[00:04:24] Louis: Yeah, it's not officially a thing. It's a, technically, it's a, master's in systems engineering, but mine, like it's a thesis and it's all around AI. And I, I do research and I decided of my subject with my professor. So it's really AI focused, but if we see the technical term I'm doing systems engineering, masters, 
[00:04:47] Amy: Okay.
And so you're a master student and a YouTuber. What do you do for your day job? 
[00:04:56] Louis: The job you'll mean other than these two or, okay. Yeah. Well, I'm a, full-time, I'm a full-time master's student and part-time, I have a part-time job as well as. A research scientist at designing Stripe, a startup in Montreal, which aims to create illustrations for anyone.
So basically the end goal is to allow a user to like, just type a sentence such as I want a picture of a. A kid that throwing a Frisbee at the Douggie and we will generate multiple variations of this picture that you could use for advertising or anything. This is the end goal and we are working on multiple smaller features and AI computer vision based algorithms.
[00:05:45] Amy: Okay. So you have your pot in all your hand in all of the AI pots. When did you get started with learning about AI? 
[00:05:55] Louis: Oh I think I started, I don't remember the exact year, but a couple of years I go at the end of my university. So quite late and yeah, it was a class, my current master's degree supervisor gave, and I really loved it and just, I knew I wanted to do that since I didn't really like any other systems engineering classes and I didn't like my internship, my internships as well.
So yeah, I think it was around the end of 2017. I think. Yep. Okay. 
[00:06:35] Amy: That's really cool. And let me ask you a question about AI now. What has been your favorite use case of AI that you've learned about thus far? 
[00:06:45] Louis: Thus far for, from when I started learning about.
 Well, too many. Yeah. Yeah. Th there's just too many. I will just say what I had, what I have in mind right now, but I think unfortunately we cannot use it, but I would say deli, just because the the image generation I go rhythm made by open the eight. Just because the results were extremely impressive compared to what we had just the day before it was released.
And yeah. I'm, well, my work is around generating images and working with images. So I really liked this one and it's really unfortunate that they didn't share the code and more about it, but I think it's really amazing. And also the recently there is, there's also the deli route or like a restaurant version of that.
Seems to be just as good. So it's quite exciting. And yeah, I'm eager to try to implement them myself and maybe improve them. Oh, okay. 
[00:07:45] Amy: So I haven't really heard about this yet. What exactly does it do or why is it impressive? 
[00:07:51] Louis: Basically it, it uses another model called clip to to try to understand and link, text to image.
So basically it will and code, it will take a sentence and take an image and measure how similar both are. So if, yeah, I think it's clear. So yeah, we use this, it uses this model to then generate an image from text. So this is what it does. And. It works with pretty much any text. Like you can type a, and that the most popular was a, and the advocate, the wheelchair, and it just generates like multiple variations of chair in avocado or green shapes.
And it really worked. It worked really well. Oh gosh, 
[00:08:39] Amy: fun. I like that. Okay, cool. That reminds me of that. Silicon valley episode, have you watched that TV show? Okay. It's like they develop an AI that connects to your fridge. I forget exactly what happens, but it will tell, it would tell you, like when you ran out of food and it would, they would also like.
Scan items and with their camera so that I would say what it was, but the guy who developed it just made everything hot dog. So no matter what you're pointing out, it was just hot dog. Okay, cool. So let's go. Tell me more about starting this YouTube channel. When did you start to get some traction with it?
In terms of growing your following and success. 
[00:09:26] Louis: Yeah, that's a hard question. I'm not really into the numbers and instead of sticks, like I like statistics since I'm into data science, but I don't really like tracking my own statistics just because it's like in the short-term it's quite discouraging.
You, you often see plateaus and even downs. I think that it really started when I stopped using the text to speech application I was using and you and a microphone and use my own voice. Instead, it started working a little better from there. And then. I don't know the dates, but a bit after that, I also started sharing articles on medium, which is really a great platform for sharing articles.
And it also helped me reach a lot more people into the data science world with towards AI with whom I'm affiliated now. And they are really they're awesome. So yeah, this is when it started. Getting a bit more views and also another big, good jump was around the end of 2020 when I shared like a yearly review on the GitHub and on my own article.
And a lot of people shared it and loved it. So this drag a lot of people to my channel as well. 
[00:10:52] Amy: And then you also started a discord channel too, right? With to connect with your. 
[00:10:58] Louis: Oh, yeah. Yeah. W well, yeah, this is not really, it's not to connect with my community. In fact, it's really to learn AI together, which is the name of the distort channel.
And it's just a community. Anyone can go to ask questions find teammates find jobs and. Just talk to other people in the field of artificial intelligence, but it's not related to my content or my videos at all. It's just, I really wanted to create a community around AI because before I made this one, there was only like Facebook groups and Reddit separately.
And this is not well to me, this is not really communities. Like you cannot just chat with anyone and jump in a call and just interact life. Yeah. And I started this in the May, 2020, and I don't know why some, well, I first started sharing the discord through my own platforms and I was looking at.
And other people that were interested in helping others that are now admins of the server and from there people joined and discard made us an official partner of discard. So official AI server, and this helped a lot to drag other people in because we were now feature. On the search page of this garden.
So yeah, now it has 20,000 members and it was pretty much because of the first few people that join and were willing to help others. 
[00:12:38] Amy: Yeah. Oh, that's great. And it must be nice to you be able to use a platform to connect with people in the industry too, especially because in the past two years, three years we've been having the COVID season.
And so you can't really meet people in person. So it must be nice to have that platform to connect with people. And I guess you have people in there from all over the world 
[00:13:04] Louis: to, oh, yeah. And yeah it's amazing. Especially for a meeting. I'm doing both my master's degree and work from my home.
So I'm just always alone and yeah, it's great to connect and exchange and I've met so many people. It's amazing. And I actually we actually hired three people from because of the discord server, so it's and they are awesome. 
[00:13:27] Amy: Wow. That's fun. Okay. I have some questions for you about your YouTube channel, specifically some selfish questions for me, because I would like to know what you think your secret to YouTube successes.
[00:13:43] Louis: I don't think I have anything special, especially because I'm not that fluent in English. So it's like I do mistakes and to me that's normal. Everyone does mean. And I will say the secret is consistency just because, but I actually said that in in the notification slugging we had, and I don't remember his name.
I think it's some, but the person asked Well, he asked how I succeeded, but mainly he said that he also tried and failed. And to that, my question will be what is failing to you? Because I would assume you, you cannot really fail you. If you just keep going, I'm convinced you will succeed.
So you basically gave up but. Fail at succeeding on YouTube. You just didn't give it enough time to, to work out for you. So 
[00:14:39] Amy: what do you think is enough time? 
[00:14:44] Louis: Oh yeah. Yeah. That depends. Yeah, it really depends on the person, but if you love what you're doing and if time is a lifetime, so you can just keep on doing what you love.
But the main goal is just to do what you love and you don't need to even check the views or whatever, if you love sharing and trying your best to explain something or just to show something. It's worth investing your time in doing it. Just for the sake of. 
[00:15:11] Amy: And I want to talk about YouTube strategy too, and the different elements of a YouTube video, because as somebody who needs to upload content, you, there's also things that you have to optimize, right?
Like you can't just bang your head against the wall and continue to pump out content. That's not gonna work. So what is it about your YouTube videos? Do you think it's the titling or the description that you put in or the editing or, 
[00:15:38] Louis: yeah. Actually, I think I'm quite bad at all of these. So especially for the title and thumbnails, like I D I just shared the most simple one that, that explains what happens in the video or what is the video about, but I'm extremely bad at making it click worthy or.
Just hi, high quality style. I'm just, I don't know. I'm not a designer and I think I should invest in to that, but yeah, right now I'm just focused on trying to improve my skills, my explanation skills. And to me, editing and views are not the priority. I'm not really focusing on that. Of course, I'm trying to also improve editing just to make it more fluid and nice to watch, but I'm also sticking to the basics and sticking with what works and feels.
[00:16:31] Amy: That is very refreshing to hear also, because as a marketing person, I am all about the marketing analytics. So just you're saying, just do what you want and everyone else will follow. 
[00:16:47] Louis: Well, everyone else will follow. The people that are really into it, we'll follow in and that's what matters. Yeah. 
[00:16:55] Amy: Okay, cool.
So how else do you connect with your community and followers?
[00:17:00] Louis: Well, it's obviously through the discord channel the most, and also I share a lot on LinkedIn, Instagram and Twitter. I think that the best place is LinkedIn because I'm actually chatting with a lot of people. And I also met a lot of them from LinkedIn and it's really a great platform just to, to discuss and share.
Yeah, I would say, do 
[00:17:25] Amy: you share your articles and your videos as well then? 
[00:17:28] Louis: Just my videos when they are out, but nothing else I share. In fact, I share what I want to share. It's, it can be pretty much anything. For example, today on Mondays, I have a weekly poll about AI that are, that I share on this card, but I also share it on LinkedIn and Twitter, and it's fun to see the differences in communities.
The votes, like some, sometimes it's really different on LinkedIn versus this guard. So yeah, I share, I also share memes and other stuff about AIA 
[00:18:05] Amy: yeah. Okay, cool. That's fun. And Oh, I had a question or not. It completely forgot. All right. I'm gonna change directions then. So what do you have planned for the future of what's AI then?
[00:18:22] Louis: Oh, no.
I, the only thing I can say about this is I will keep doing it. What I want to do. So I cannot be more precise than that. I don't know. I think I will keep on explaining new papers and new research because I love it. And I still want to stay up to date with research, but I'm also working on many other projects and a lot of more ambitious projects, which are taking a lot of time.
And. And yeah, I really don't know where this channel would go in my blog as well. 
[00:18:59] Amy: And what about when you finish your master's degree? Do you have any thoughts on what you might do after that? 
[00:19:06] Louis: I think I will just come back to the same answer,
but I can be a bit more precise. I'm currently. I'm not sure if I want to do a PhD, but I'm currently applying for PhDs just in case, because I think I want to do a PhD, but I'm not sure yet from what I've heard and thought with others, but still I'm applying just because of. Well Yolo you, I don't want to miss out.
And I will just deny if I get accepted and I don't want to do it anymore, but yeah, I'm hesitating between going full-time at design Stripe and a bit more time on YouTube or going into a PhD, which yeah it's quite different, but yeah, I really don't know. 
[00:19:56] Amy: You really value higher education then.
[00:19:59] Louis: Not that much. It's just, I love the current environment I'm in. I love working with a professor and a team and trying to do pure research. And I think I prefer pure research as a research into industry, but I'm not sure. So this is why I'm his day thing. But if I can manage to find. A great team for the PhD and a great professor.
I think I could really love it. Love the time I'm there, but also it's four to six years. So it's a long time without a lot of revenues. So it's a bit, it's a hard decision, but yeah it's not that I think a PhD is necessary in AI, I think. Not at all, but. Like I've always wanted to be a researcher or to do something that is, that, that is hard to achieve.
And I think a PhD might be on the right path to, to this. 
[00:21:05] Amy: Yeah, that's super cool. Okay. Now I remember my other question. So I'm gonna circle back to that. Take me through your daily life. Like what, how do you structure your day and what do you usually do in work? 
[00:21:18] Louis: Oh, good. It's I don't have a stricter. I just never ending to-do list and yeah, basically I have, I still have an, a paper agenda where I just write down what I need to do today and what I will like to do as well.
And I tried to do the most. I can and yeah, except going to the gym every day and running, I would say that all my days are different and the rest is just, I feel it with working and doing U2 stuff and research something. Okay. 
[00:21:58] Amy: Well, wait, do you like get up at the same time every day? Do you go to the gym at the same time or is it just like whatever you feel.
[00:22:08] Louis: I know, it's just, when I afford a gym, it's when I feel best I will go training or when, or if I have a lot of meetings, I will just go whenever I can, but I don't have a fixed time. And for my waking hours, it's I'm just having fun with it. I'm actually I have a random number generator between five 30 and 6:00 AM.
And I'm just playing with it with random. Like I'm waking up at 5 37 or 5 31, but I'm also always waking up earlier than the actual alarm. So I'd say I wake up pretty much around the same time, like five 30 every day. Except when I do something late at night. For the last night where I really aim for at least eight hours of sleep, because I think sleep is extremely important.
Even if you need to work a lot and be productive, I'm you need to sleep at least seven hours a day. And yeah. 
[00:23:09] Amy: Yes, I absolutely agree as well. Do you think that you thrive in the chaos? 
[00:23:15] Louis: I. I totally don't know because I, for me I'm quite structured, but yeah, maybe my guess is structured.
I, I. Structured chaos. Yeah. Yeah. I don't know. Maybe 
[00:23:31] Amy: it sounds wild to me like the, even just the alarm thing. Well you never know when it's going to go up. I love that. Well, that's 
[00:23:40] Louis: just fun, but yeah. But I also have I'm tracking my time. For example, I have I'm using , which is. And Adam to track your time on what you are working.
So I know for example, I'm doing 40 hours of masters researcher weekend, et cetera. So I know where I'm at, but I don't know where what I will do on the day to day. Like I more I'm tracking on a weekly basis. Like I'm releasing one video and I'm working 20 hours for designing strive for the hours for you for master's degree.
Like at least this and more if I want to. And yeah, other than that, the days are pretty random. And also I have a rule that I don't I don't do anything I don't want to do after 6:00 PM. So after 6:00 PM, I just do what I want, which can be a, watch a Netflix TV show or our work if I really want to work.
And it, it often happens and yeah, I it's been doing great then. I love it for now. 
[00:24:49] Amy: Yeah. I love that rule. That is such a good rule, especially to incorporate some good work life balance into your life because yeah that's a great role. I love that. Okay, cool. What do you do? Do you have fun other than like working?
Cause that sounds like a lot of work in school time. 
[00:25:10] Louis: Well, I have fun working, so it's already fun, but yeah, I, well, I just love going to the gym. I'm just completely addicted. So I have fun there and I'm also going to the gym with some friends are listening to audio books. So it's always a good time and same thing for running.
I'm also running alone with an audio book or with my friends. So it's really enjoyable. And I also. Doing what I'm seeing some friends from time to time, which is a bit harder with COVID. But yeah, and I'm also doing a lot of sports with my friends well, not a lot, but we are recently we've been playing I don't know if you know about this, but we've been playing spike ball every week, so yeah, it's really, it's like the best.
That was created. 
[00:26:02] Amy: I have never blade, but I maybe yes, the start of 2018 or something, just popping up on the beach everywhere everybody's into Spikeball now. Okay. I got to get into this. Damn. All right. So as somebody who's very into AI, how do you use AI in your daily life? 
[00:26:22] Louis: I just, I'm using my phone.
So yes, I'm using AI because it's just in all applications pretty much. 
[00:26:29] Amy: Oh yeah. Okay. I guess so I never really thought about, 
[00:26:36] Louis: but yeah, I'm not, I couldn't really see other than that I'm quite simple and yeah. Like doing when I'm not working or doing AI related stuff, I'm just doing sports or relaxing.
So not really. I'm using the Netflix algorithm a lot and it doesn't work that well, but that's pretty much it. The AIU was the most . 
[00:27:01] Amy: What do you think about the future of AI? Because I think a lot of people like me too, when I think about AI, there's just so much of it. We forget how much of it is used in our daily lives.
And sometimes it feels a little bit like I'm a scary robot in the cloud that's coming after you. Right. So what do you think live future of AI and adoption of AI is going to be like 
[00:27:31] Louis: for the adoption of AI, I'm a bit scared because governments are a lot behind, so yeah it's a bit scary.
I'm actually like there's typically two sides, like optimistics and people that think of a singularity and the worst thing that can happen. And I actually don't know. I'm really more agnostic just like for religion. And I just, I think I will open these, say that I don't know what will happen.
Hope for the best, but yeah I really cannot say I don't mind. But 
[00:28:10] Amy: I guess like we are already using AI in our daily lives and we don't even think or know about it, all this kind of social media algorithms or, I guess like email sometimes, or like everything. 
[00:28:24] Louis: Oh, yeah. I would say in the end, I would say the AI I use the most is definitely Grammarly, which just corrects your sentences and yeah, this is amazing.
And yet in fact, I'm like totally working for for design Stripe on a series of articles where we will explain w where AI is used in our daily lives and in all industries, just because I believe that this is nobody knows. Well, a lot of people don't know that they use AI everyday on a lot of different applications.
And even some the, they thought will be like extremely simple or not related to AI where they are. 
[00:29:05] Amy: Yeah. Yeah. It just, the term like artificial intelligence feels very like futuristic robotic, that is so unaccessible that it's just not being used, but of course we're all using it. I use Grammarly every day.
That's right. I forgot about this. So many different applications. 
[00:29:25] Louis: Yeah. And sometimes when we say AI, it just means like a basic algorithm. That isn't really indebted and just statistics. And so this is why it's used everywhere, just because the term itself is so broad and we don't even know what it means 
[00:29:41] Amy: yeah.
Yeah. Okay. Demystifying AI, let's go. Okay, cool. So if we want to find you, when, what you're working on online, where can we, you look 
[00:29:53] Louis: yeah, you can look pretty much everywhere for say. On YouTube mostly, and also I have a blog, a personal blog called my name, Louis Boucher, that AI. So that's pretty much the two main.
[00:30:08] Amy: And it, you can check out Louie on hacker noon. He has over 50 articles there which is definitely, you have been one of our top and most loved contributors. And again, I will read you the list of all of the hacker noon contributor of the year awards he is nominated for that is under the computer vision category, data science.
Deep learning Facebook game development, Google, iPhone, and machine learning. If you want to vote for Louis for these awards, I will put the links in the show notes. You can also visit noonies.tech and if you make an account, then your vote counts for more. So go for the URL and noonies.tech, make your account and vote for Louis on all eight of his awards that he is nominated for.
And yeah, that's it. Stay weird. And I'll see you on the internet by 
[00:31:03] Louis: hackers. Thank you for having me. Yay.




Written by podcast | Tune in to Listen to Tech Stories from HackerNoon 2-3 times a week!
Published by HackerNoon on 2022/01/05