comet-oh-march7.mp3

[00:00:09] Welcome, everybody. Good to see you guys here. Welcome to the comet. I know office hours powered by the @TheArtistsOfDataScience Superexcited. Have all you guys here. Hope you guys had a excellent week. Thank you guys so much for taking time out of your weekend to come and hang out with us today. Super excited to have you guys here, if you guys wouldn't mind. Go ahead and put in your microphones on mute so we can minimize any background noise. Definitely appreciate that. It has really been it's been great.

[00:00:36] It's been busy on my end, but getting some things moving. We've got a lot of exciting announcements for Comet coming out this month. So just getting getting in gear to do that. I've been doing more more marketing work recently.

[00:00:50] I thought of dude, I was actually on Super Data Science podcast yesterday on AIs so that Jon brought me onto the show and interviewed me. So I was super excited about that. It's the first time I've ever been invited to Data Science podcast. I've been on a couple of other podcasts before, just like some friends and stuff. So that was really special to me because, I mean, I've never been invited to a Data science podcast and to be invited to the Oge Data Science podcast like I was the first one that is huge for me. So I was really, really honored to to speak with Jon on that show. So I'm excited to release that episode. But we obviously were talking about you and talking about our hours as well during the episode. I did not know you were writing a book.

[00:01:32] Yes, I will say super Data science is like one of my favorite. Like you mentioned, it's Oji podcast. I loved it. Had the pleasure of being on recently. And yeah, the book I'm currently writing is about bias, machine learning, going into the different ways we can technically solve these problems. So explain ability, interprete ability, but then expanding to like operationalizing ethics in A.I.. So how do you convince stakeholders to choose the less risky models? How do you be transparent? How do you word your terms and conditions to clearly state the kinds of data you collect and how you model it? So it's going it's going pretty in-depth into trying to tackle a lot of the fairness problems we've seen recently.

[00:02:21] That is really awesome. I did not know you're writing a book, so now I've got to get you scheduled and brought on to my podcast so we can talk about that, because that's something I've been thinking about a lot recently as well, because I don't know anything about ethics or machine learning ethics or anything. So definitely excited to learn more about that. Something else I've been thinking about recently, passed a couple of days, really is like, what does it mean to be a expert in Data science? What does that what does that mean? What's that spectrum look like to go from novice beginner all the way up to to expert and I don't know. What are your thoughts on that?

[00:02:57] Yeah, I think a lot of people it kind of is like specialization. So typically you are a you get really deep into NLP or computer vision. I think that's kind of how it starts. But even for a lot of people, this expertize can be in industry as well. So like people who are super strong at product analytics or people who are really in depth into fintech, I think that the gap is really about experience, about having hands on either projects or solving the difficult problems and having to talk through what solving that looks like on a team of at least one other person, I think really helps you get to that level of expertize. And I would say just having enough and because the field is growing so quickly, it's like having enough knowledge to be able to be helpful, maybe a little bit to be dangerous, but still not knowing everything. You can maybe know 60 percent about everything that's the core of computer and probably be a computer is an expert.

[00:04:07] And there seems to be like a difference in the way a beginner approaches a problem versus how an expert approaches a problem. I would say beginners very much are looking for a recipe, a formula to do this, then do that. And then if that doesn't work, try this. Try that. Right. And they're very much bound by rules. At what point in your career do you feel like you started to deviate from falling recipes? And how did you get to that stage? What did you have to do to get comfortable enough to recognize that? OK, this is probably somewhere I can break the rules?

[00:04:41] Yeah, I would say it came from a lot of the insights I gained from trial and error. So a lot of my modeling has been in like traditional Emelle. So doing things like decision trees, generally Glenns and then putting those into production. So I got to that point where it felt like I could start breaking the rules when I started to see like mismatches in. And I expected so you run a model, you expect a certain result, and you find things that deviate from that. So trying to understand, is this a flaw in my Data? Is this a flaw in my assumptions going through that process several times, kind of open the door to deviating from what a trait like a stereotypical training process would look like for a model? And I'm like, oh, I can do a lot more things. I can do a lot more feature engineering, a lot more transforms before trying to create a different creative model.

[00:05:42] And this is something I noticed with with people that come on to the podcast. And officers like this is like we'll have somebody ask a question and then typically almost like a gut knee jerk reaction is it depends. Right. Because there's always the context that needs to be taken into consideration when you're working on on any problem. Right. So when you're I feel like when you're working in like a context free environment, when you're looking for answers that are context free, to me that signals maybe might be in that beginner phase. But once you know enough to know that you don't know enough and that you can start asking questions about context, I think that kind of is slowly shifting you up towards that spectrum. But that's something I've been thinking about and reading this book right now. Pragmatic Thinking and Learning by Andrew Hunt, who also co-wrote the book Pragmatic Programmer. So I'm interviewing Andrew Hunt this next week, I think next week at some point. And so it's just really meeting, thinking it's really, really good because they recommend it to everyone who is, you know, interested in reading books. But, yeah, thanks for for for helping me think through that. I appreciate it. So, guys, we got questions up in the queue. We'll start with Cristoff, then we'll go to Yosh and shout out to everyone. Join in.

[00:06:54] Happy to see you guys here Cristoff go for it Tirupati how great a message. I sent you a message today asking him if it would be fine to us to talk a little bit about mentoring because last week you mentioned that you had some people to get possibly better results during job interviews and I feel like I also need a mentor right now. And I just want to do a couple of questions about it for, for instance, what is actually mentoring, what is meant in this world.

[00:07:41] And I mean, what does it really mean to be a mentor or a mentee or and if anybody can become a mentor? If not, what could you have to be to become one?

[00:07:55] That's pretty pretty interesting question. I think I just accidentally found myself becoming a mentor somehow. And I joined Data Science Dream Job as a student myself, like three years ago, back in twenty eighteen. And Kyle McHugh is probably the like. I'm not even like I say this. He's probably the only mentor figure that I've had in my entire life, throughout my career. I very much had to figure shit out by myself all the time. And that's very unnerving. It's not easy. And so I joined the industry in twenty eighteen and it's like just essentially a mentorship coaching platform for people who are breaking into the field of data science. And I took a lot of the lessons to to heart, made them my own synthesize them. Right. So I did a little bit of imitation assimilation and some innovation and then kind of took what I learned through Data Science Room job and applied it to my own job search process. And I guess Kyle was yes, he likes my results and liked kind of how I was helping out the community just brought me on board as a mentor, like there was nothing that I was planning to do with my life. I really like I didn't didn't have that in the cards. And so that's how that happened. So that's how I became mentally stream have. So I don't I don't have, like, mentees that I just like like scout out and see how you come and be my mentee. Right. It's mostly just two days and screengrab that I do this. So that's how that came about and make phone talk about this.

[00:09:15] And I seem to have definitely can can send you some more information on that. But there's another part of your question, which was can anyone become a mentor? I think, well, what is a mentor, I think is just somebody who is maybe like a few steps ahead of you in the process. I think those tend to be the best mentors, is somebody who has been there, done that and maybe might be like three to four steps, five steps ahead of you and can kind of tell you what to watch out for, what to do, what not to do in that respect. I guess anybody can can be a mentor, but I don't think that's like a relationship people should take lightly. I get messages. I get literally like a hundred messages a day from people like please be my mentor. I just ignore them all. I just don't take on, like, random people to be my mentor. And less obvious because it's a time commitment and like trying to like, sit down and like, give my time to somebody. That's time that I can never, ever, ever get back. And I've had students or people reach out to me, want to be my. Maintain whatever and give him advice, and they just ignore it completely, and as I do, I gave you like four hours of my life that I can never get back. And you're not going to listen to anything I told, you know, but I'm going off on a tangent. Sorry, I totally flipped this one over to you as well. What does it mean to be a mentor?

[00:10:25] Yeah, I think, like you mentioned, someone who is just ahead of you in the process. And I'd qualify that by saying they've made enough mistakes and learn from them to be able to do so. In my experience, I mentored on the sharpest lines, which is a Data science mentorship platform as well as mentor crews. So what I gained from a lot of that was my job was to sit down with people, help guide them through projects, so they'd come with a project idea and say, I want to predict housing prices. We sit and go through the features. They kind of explain that project to me in the way they would an interviewer. And then I'm able to suggest, OK, we tried stepwise regression. Have you tried doing X, Y and Z and through the whole like end to end NLP project. And then I've helped with like Inari Prepon. So like mock interview questions. I was able to help you on this project, but explain it to me like you've never met. And those what I that's what I would say were my biggest responsibilities. And but like I mentioned, it is a really big time commitment. So both of those platforms just kind of help bridge that gap because it's really difficult to kind of organically form a really strong, like, mentorship mentor mentee relationship if it's not structured in some way. So if it's something that you interested in, I'd say comes from what you really need to be a mentor is need to have made enough mistakes to then guide someone else on how to not make the same mistakes you did.

[00:12:16] And there's a couple of my mentees inherited from Data century job is here, is here as well. So these are people who had joined the dream job as well. So, I mean, I guess they could speak to my effectiveness or ineffectiveness as a mentor. But for me, like when I think about myself as a mentor, like I'm I'm extremely tough on myself. I got very, very, very rigid requirements of myself. And I try to be more tolerant for my mentees. But I do like I'm kind of like strict in a sense. Right, because I do want to see people excel and really live up to their full potential and realize their full potential. And like my personal mentorship philosophy is like, I can't get you there by being soft and fluffy with you, that there's always going to be more rigid with myself. But I'll be just as tough on my mentees out of jail or Siwash want to comment on on what their says dream job is like, what the experience was like.

[00:13:16] Yeah. So for me, actually I have I have not really make use of the Data Science Dream Job mentoring site. I mean the that portion of it yet because, you know, trying to learn a lot of other stuff before I come into the mentoring part of it. So I can't speak too much about it. Sorry, but I've been attending Harp. It's mentoring for the Friday and the Sunday, so that has been my go to the moment. So. So yeah, I can't speak for the year yet, but I definitely like so far, whatever the setup that they have, that's been very good for me. At least I have some sort of a guidance to how to reach my path as a data scientist before I was all over the place, know there was no GUI, nothing. So no I was just doing this, doing that. That was just all over. So I needed a structured program to kind of take you step by step. I need to do this first and do the next module first and then the next. And so it's really good right now. I'm like in the face of it, like learning all the different topics and subject and hopefully all of that is done.

[00:14:20] Then I will go into the mentoring session when I have a project ready to show for. So I think the mentoring part is definitely valuable. I think when you have someone who's ahead of you like what I said, you know, they've been through it, they've done it, they've made the mistakes. And, you know, and they can help those who are trying to get there for sure with the experience of the staff needs to super, super helpful. And I wish I had this mentoring earlier on in my life, but now it's my journey to becoming a Data science. I know someone to say, hey, don't do this, do that, you know, and that what I've been super helpful. I would have saved me maybe years of doing stuff that maybe didn't matter, you know. So I think that was the thing, because time is something you don't have. Right. And it keeps on moving, keeps on moving and you can't catch up, so. I think this time components of valuable. Yeah, I'm I'm all for mentoring and like I said, I wish I had this earlier and years ago, I would have really, really helped me up and stuff.

[00:15:20] Yeah, absolutely. I feel the exact same way. It's funny that you you go to all my free events, but not pay for sure.

[00:15:28] I know. Yeah. Yeah. I'm not ready with a project. I want to get into it pretty soon once I finish all those, you know, the I think that the set up like this does for statistics. That's what I did this for Programming's. I'm trying to go through the modules and that production is taking a lot of time. I feel it, but it's very valuable. So I must say.

[00:15:48] Yeah, and I wish I had something like that as well. And I think the biggest thing about having a mentor is like I've made a ton of mistakes, both in job search process, on the job in my life and my career. And you don't have to make the same mistakes. I'll tell you what I did. And you can avoid doing the stupid shit that I did on my path to get here. So that's a huge plus as well as Isobar shows unmuted. If you want to chime in, too, go for it. I think that everything that David is also at yesterday's student as well. So there's a lot of guys here today.

[00:16:16] Yeah, I definitely I'm not sure like if you were hard on me or something like that, I always and like meekly it looks I need one of your officers for the last two weeks and you are preparing for a drama school. You didn't have your office hours since I was missing those sessions. Definitely. So I always learn something when you stop. So I always like to be on what it is like a and I like the idea of I said whatever it is, I always get something to learn from it.

[00:16:46] It's like being on your office and at the very least you get to hear me just ramble on about things for minutes on end, which can be entertaining. Um, Cristoff, if that answer your question or not.

[00:16:59] It does, although I have a follow up question, if you don't mind.

[00:17:05] Yeah, sure.

[00:17:06] And so because you're using like platform art, there are also there options like person to person? I mean, could I approach a person with a suggestion? I mean, may I pick someone or it doesn't work like this. I say, yeah, yeah.

[00:17:32] So like I mean, so just boring. The talk we're doing about Data screamed out, that's all. Just officer sessions that are a bit more intimate than what we have here, like five to ten people at the most. But if you are somebody who is looking to find mentorship from someone else, I think the absolute wrong way to go about it is just to message someone and say, hey, can you mentor me? Because there's absolutely nothing in it for that person to mentor you. Obviously, people are good. We're all good people. We want to help people. But to ask somebody to just give up their time for to a stranger for no purpose, like, that's a huge ask, I think. Um, so if you can find something that you can help them with. Right. Like let's say for example, let's say you approach me and say, hey, are brick like can you be my mentor and be like, sure, what how how's this going to be a mutually beneficial type of arrangement? Right. If you can help me free up time somewhere else, that then gives me more free time to work with you, then that's a win win.

[00:18:28] A positive sum game. Right. So if you're going to approach somebody to be your mentor, then try to make it a positive sum game somehow. Does that make sense? So let's say you reach out to I don't know, let's say you reach out to my friend David Llegar and say, hey, David, I'd love to be your mentee. I know that you do these videos. I've got particular skill in editing video. I'd be happy to offload some minor video video editing from you if you are able to help me through my job search process or something. Right. Try to make the offer mutually beneficial and make it a positive sum game for someone, because if you approach somebody to be a mentor and it's a zero sum game, meaning they have to give up their time in order to help you, I don't know if that's the best way, but then again, that's just my preference and the way I think about things, ideally, what do you think? You might be a far more kind and generous than I am.

[00:19:17] But I would say, first of all, that's a good place to start the tactic. I would urge not you just kind of the very general. I am looking for a mentor, but I would encourage you to find you might have to do like a Twitter keyword search. But there are some people who are doing like office hours similar to what we do. So a lot more one too many kind of mentorships, especially if they are someone who's got like thing experience. It's going to be a lot easier for them if they're if they're already doing like an office hours to try and get one of those slots, then getting like a one to one kind of mentorship. So I would also suggest kind of being open minded to some of those small group options as well.

[00:20:06] I'm hoping that answer your question because of. Definitely, and also answer my next question, actually, I have a question how to make it mutually beneficial and make it a win win situation and you just describe it. So thank you very much. That was I was very, very helpful.

[00:20:26] I definitely think the more we learn to play positive some games, the better off everybody will be. So if you can, you can try to find ways to to make somebody else's life easier to get you ahead. Go for it. Positive sum game. All right. So cute. Up on the question list. I've got Josh in Asia than tour, so we'll start with Josh. Is Josh still here? Yeah.

[00:20:46] Hello, everyone. My name is Rudd and I'm currently doing Bachelors of Science in Applied Statistics and Analytics and I am very first time attending the Harp. I saw the post of FBAR on LinkedIn and I didn't attend. This is my first time. I hope I will enjoy this and I am doing a side-by-side quantum computing course. All right.

[00:21:12] Well, I hope you enjoy it as well. Thanks for coming. We have the every Sunday, so feel free to come by. Added to your calendar. They'll be here. We'll be here every Sunday, at least until May. So I don't get to meet you. If at any point you got a question, just let us know. We will then move to Asia.

[00:21:30] Hi, everyone. I had the same question, but it has been answered. My question was also on the mentors. But but my question is, how do you know what skill you're going to gain? Do you look at areas where you're weak before you approach a mentor, then tell them these are the areas I'm weak? Or do you just approach it with an open mind and hope to learn anything they have to offer?

[00:21:50] I would take the opposite approach out approach somebody just because they have a specific skill set. And that's the skill set that I wish to learn. Right. Like, that's kind of the approach that I would take because you might have a goal to get better at computer vision. Let's just say, can we talk about that earlier? But if he came to me and asked me to mentor you in computer vision, dude, I'd be completely useless because I know like zero things about computer vision. But if you came to me and said, hey, look, I'm looking to up my statistics game, then yeah, maybe I might be able to help you. Or, you know, you want to develop strength in classical machine learning, then, yeah, I could probably help you, but you have to make sure that whoever you're reaching out to, you target them for a specific skill set that they are known for, if that makes sense, ideally. What do you think?

[00:22:36] Yeah, I think especially if you are hoping to specialize, that's a great route to take if you want to be a little bit more general. I think if you focus more so on the kind of project you want to complete, so maybe you are looking for someone who has skills, computer vision for computer vision project you want to do, making sure to target those people. Well, and it's kind of like the recruiter tactic, like I'm sure this is fairly common, but you're trying to get into analytics and you're getting jobs are web developers and you're like, well, that's not even close to what I'm looking for. So trying to target your mentor is the exact same way. So if you're into natural language processing, trying to find people, find people who have written papers on it or who have talked at conferences about it, you can start seeing those as good tools for different kinds of mentors you can potentially reach out to.

[00:23:36] Thank you so much.

[00:23:37] Yes, it did. OK, great. Thanks for the questions. I was super happy that you guys are asking some awesome questions here. Tor is coming here with a he's got a dilemma tor what is your place.

[00:23:51] But technically right now, just to close, this is kind of like a little bit off topic on the mentorship. But right now I'm developing a source or a concept and this is now going to be and I'm working on the analyzing and estimating the users or subscriptions that are going to be sold. And the challenge I have is that the subscription is kind of like a twofold one, which is paying at one, which is free and with the free one.

[00:24:28] What you want to do is to allow users to access the tool and share and by sharing that will now generate new accounts. It's an automated process that's technically attracting them and trying to attract new users. The problem of having is that when I'm analyzing it and making a model now to calculate this, I get into a circular reference problem in Excel because I'm saying that one user will generate three new users, for example, every three months. Each of those new users will again generate three new. So I can bring it back into the new calculations. And I'm trying to figure. A way to calculate this either through a formula, if there is some sort of a 100 percent exponential form or another type of sort of type, and to bring that into the model they're creating, or is there a way round the problem of this regeneration into the new calculation? I don't know if that makes sense.

[00:25:30] That's going to take me a minute to digest. But I mean, my first initial thought is like, can you get a database up and running and work from there? Because that's like using Excel sheets might be a bit of a chance.

[00:25:44] That's the challenge with Excel. It is clearly a challenge that when and how how I'm actually budgeting new users is by putting in a period I'm spending four to five years, but for the first year I'm kind of doing a monthly input.

[00:26:02] And then for year to year I'm doing quarterly year three and doing half yearly and then four and five year. So I just plug in individual incremental values per period. Now, from that, I would then have a second table where it's now calculating in the column called Conversion Rate, where I went because I had different subscription packages. You will have an upgrade. I'd say 10 percent of users will upgrade to the next package, but then the tab. But then I get into the issue of this particular sharing where they are using the tool and sharing the report with other friends and family or workers, etc. And every time that's done, it creates a new account automatically, which the receiver will see. Now, in my model, I can't then bring it back in to the conversion table because then I get the the circular reference. So the conversion table. Then of course, all these free accounts are created. A certain portion of them will then be converted into paying customers and then they will be brought into the model so that they will be upgraded as well. Because that timeline is that you get a new user discount within three months. They will then upgrade to the next level and to the next level again after 12 months and so on and so forth. And the problem is that this waterfall effect is just not possible to do an. So the question I really have, can it be done? That's one I'm not an expert, but in general. And if you're doing it, what type of tools are you using? How are you doing it statistically? Is there a formula I can use? Is there some sort of pathology I could bring into this? That's my dilemma.

[00:27:53] Yeah. So ideally, I think this might be something you probably deal with that comment, right, giving you guys business models and very similar to this.

[00:28:00] Yeah, I've definitely worked something with something similar along the lines of this. So in that regards, it can be done. In my experience, what I used is a combination of tools that we were using, I think was amplitude for like product metrics in combination with ultrarich. So that's just a UI tool that you can wrangle Data with, clean data with create models within. And then we actually had that exports similar types of tables that you're working with to excel sheets that we ended up using for dashboards and reporting. So I would look into. I'm not sure I know Ultragaz is mostly pain tool, so I'm not sure if there's something along the lines that is free, but that might be helpful in that it allows you to get data from different inputs. And if you need to make transformer's within that, you can do that as well.

[00:29:04] And just kind of beyond the tool, maybe the key word that you want to search is probably some combination of cohort analysis, conversion, attribution modeling. Those might be things to look up and try to see if you can get your stuff in a database. Kind of feel like that can probably make things a lot easier for you. Yeah.

[00:29:25] So it's an interesting thing because when I was calculating it today and I'm working on it today and I've been working on it for the last three days, you know, it started spinning and you just went through X amount of trillions with a need in front before I decided to stop. So but those I'm going to have to come back and listen to this again. But if you could type in just those references that you gave in the text and then I can use that as a reference and I'll do a little bit of research up there.

[00:30:00] Yeah, definitely. So just take it out. As I'm saying, it's a Schallert analysis. So Riphat. And then. Virgin attribution modeling, those might be helpful in this situation in terms of keywords to look at and then I mean, you can probably find resources in how to do it with Excel. But those are the two concepts I would look up in this case.

[00:30:24] And, um, any old concept that might be useful here, not off the top of my head, I'd say cohort analysis mostly covers it.

[00:30:34] Yeah, hopefully that helps men.

[00:30:37] I'll start there and then I'll bring it back up and we'll hear more from the podcast. I just hope we're not off topic here if I'm asking this question.

[00:30:46] So there's the thing there's actually no topic to these things. This is driven entirely by your questions. So next up, I got I got you in the queue and I got ushe again and then I got Davran. Also want to take a minute to shout out my friend Natasha Kapoor, who I have not seen in years. We used to work together way back in the days in Chicago that we haven't seen in like eight years, almost.

[00:31:09] Heyburn I good. A lot has done. A lot's happened since then.

[00:31:16] Now you dream big and you're doing big things in the business as well.

[00:31:19] So that's I mean you're the expert because you're hosting Data sales.

[00:31:26] I mean, this is amazing.

[00:31:28] This is amazing. I just filled my way up somehow and this happened definitely not an expert, but.

[00:31:34] Yeah, and you can I just did a brief introduction and I'm all right. Or I'll just wait until the other answered. I'll go for it. Go for it. OK, yeah.

[00:31:43] So my name is and Sasha before I graduated the Masters and Data Science back in twenty sixteen. So back then there's just a few programs that were available and I ended up choosing St. University in New Jersey, the director of the program. I was really impressed with his background and he developed the whole curriculum, his genius, Professor John.

[00:32:11] So he came from MIT, Caltech, and I mean, he was just very cutting edge and he sucked a lot of my blood.

[00:32:20] But it was it was one of the best time I've had. And I wish I could go back to academia, but, you know, maybe in the near future. But I do feel like it's just the competition for Data science jobs. It's just gotten enormous. And yeah, I'm kind of intimidated with people and interviews. Some here, too, you know, kind of I have a lot of brushing up to do on the theory part of machine learning. So the supervised unsupervised algorithms also in these sessions can run like just PR. You know, I did this natural language or text mining project for an insurance company. And of course, you have to kind of start out with, like building your logic, the algorithm. So maybe one of these sessions, could I have someone like kind of give me feedback on how to make the algorithm more efficient or maybe like a certain function, maybe get a reduced amount of code. So kind of like a GitHub concept, but like that definitely.

[00:33:40] Just as long as you're comfortable sharing the stuff, because these are all posted to YouTube. But I'll post to the podcast if there's any confidential information that you don't.

[00:33:49] Yeah, I was just going to use dummy Data.

[00:33:54] Yeah. So you can definitely be here. Like I know nothing about NLP next to nothing just because I haven't really worked in that space, so I don't know how helpful I'll be. But there's a bunch of people here as well that might be able to help or on Fridays there might be some. I know for a fact there's NLP people that show up on Friday. And regarding the interview process, check out the office. Our session from last week. We actually spent a fair amount of time talking about how challenging it is to pass these Data science interviews. I think you'll enjoy that session yet. It's not easy. I still get constantly rejected from job opportunities. Just, you know, it's it's part of him and it's.

[00:34:33] Yeah, yeah. My self-esteem goes down to the ground. It's like, oh yeah. But then I push myself, be like, hey, like, what else?

[00:34:44] Where do I need to improve, you know, what sort of courses shitake or you know, I read up on. But yeah, it's just job or job rejections are hard to digest.

[00:34:57] Yeah. Yeah definitely. Take a look at the session. I think that the day it was I posted it just on Thursday on my YouTube channel and we go deep on on on that topic. I think you really enjoy that. But Getsy again, Mansmann, spend literally like almost a decade Danti World is Harp.

[00:35:18] I'm also looking for a mentor like yourself. So any feedback?

[00:35:24] I know this was discussed early in the session, but so this guys, if you guys are interested, here's dsdj.co/artists give you guys 70 percent off. If you're interested in checking it out, go for it. See what it's all about. So, yeah. To check that out. Sure. And then obviously, like, you know, we got this session and then Friday session as well. And, you know, I'm happy to to help you guys through during those two time points as well.

[00:35:48] But is there a difference between the two sessions, one on Friday and the one on Saturday?

[00:35:53] They're just just different times, I guess. Yeah, but they're exactly the same the exactly the same amount of awesomeness, that is for sure.

[00:36:03] Cool.

[00:36:03] Awesome. So, yeah, it is. Yeah.

[00:36:06] My question is, I know my my end goal is to become a data scientist and I know that it's going to take some time to get there. But I like to know what job growth can I get in now? School Data relates to touch upon different Data aspects, but not as high as Data scientist. What will be some of the roles that I can try now, given the fact that I've got some some projects, I've got some no skills, tech skills, nontax skills, how can I get into a company that has a role that is has Data flavor in it? So I'm kind of I've been looking through different job postings. They all look very confusing, something I'm asking too much. Too little. So what will be the first step? What are some roles, namok roles that that can finally getting your foot in the door before becoming a data scientist?

[00:36:58] Yeah, yeah. I would say, first of all, I'd still apply for Data science roles. I just apply for them and just see get feedback from the market as to where they think you fall. Right. That's always important. Like for example, like I just randomly applied for it. Like I'll spend one month applying to nothing but CEO level rules just to see if people think I could be a CEO. Nobody thinks that yet. But I mean, obviously, Data analyst is one thing as well. But the weird thing is that companies will have all sorts of weird name for positions that when you look at the job posting in the job description itself, like that sounds like a data scientist, but maybe the title itself will say, like technical analyst or reporting analyst or things like that. The one thing I like about I mean, there's many things I like about LinkedIn, but the really cool thing about LinkedIn job search feature is that you can search by skills. And if you search, if you search by skills like we can do a quick example, actually, if you work. Yeah. Let's pull up LinkedIn here. Um, well, Odali has disabled screen sharing, which is can you can you make me the host. I really want to check my desk. Yeah. So on LinkedIn. So for everybody listening on the podcast, watch YouTube for this part. But on LinkedIn, if you go to a job search and you'll see if you click on job, you'll say search by title skill or company and you can actually search by just putting in like maybe like Python comma SQL, comma, data visualization.

[00:38:26] And then just look at the variety of job titles that come up. So if you if you guys do that right now. Right. So you pull up LinkedIn. I don't think I can because you're measuring it. No, not yet. Um, don't worry about it. Well, we'll move past the screen, but yeah. When you go to LinkedIn, like for example, I'll do live and direct right now just about showing you guys I want LinkedIn with the jobs. I'm going to search for your next job and I'll type in python. Com a SQL. Com a Data. Right. And then I'm looking in this case I will change it to worldwide. And like some of the the titles that that come up, like they don't say Data analyst or data scientist in that one is loyalty analytics analyst, data analytics lead, um, business analytics lead. And I look at the title and it's like says Business Analytics Lead. But when I look at the job description, it says who you are. Extensive data analytics, experience, expertize and business analysis demonstrate skill and ability in Python or SQL. And it's like, oh, well, that sounds like a data scientist type role to me. So try to search by skill rather than actual job title, and that will help you get exposure to jobs that don't necessarily have that title. But when you look at the contents of that rule, does it's actually in line with what you're looking for, ideally. What are some interesting job titles you've seen in your search?

[00:39:48] Um, there are a lot. So pretty much anything with analyst in it. So it can be business analyst. There are also titles like Analytics Engineer, which I don't think I would have initially targeted, especially when looking for Data I asked jobs. I've also seen a lot of organizations have Data scientists really under the Data engineer job title, and that's one that, while kind of adjacent, can have a really strong programing background. It's easier to step into a Data engineer role and then maybe step into a Data science role, if that's what you like to do. I've also seen business intelligence really mean Data science, despite the fact that they don't really mention Data science in almost anything under analytics. So there are a ton of job titles and that's why it's really hard to search for them that way.

[00:40:50] Yeah, if I could kenen. So the VA like the business intelligence jobs from what I've seen or have experienced, which is that they're looking for someone with little experience and then also the by visualization tools like powered by Tablo. But I still like those tools Data science scientists position would have anyways, so I can get all kind of overlaps. Right. So the most common tools are that I've seen on job descriptions are definitely SQL are looking for t t SQL or no SQL for your storage like DBMS tool and then you're actually the tool that you use for model building is the popular one is Python. I mean I use R but I don't think it really matters. And then your Data of this tool, the most common ones area. Yeah. And power by like here's an interesting one.

[00:41:56] Bit like I just saw, I pulled up, I typed in skills that I was looking for in this job title is called Strategist. That's just the title that they're hiring for strategist.

[00:42:06] And I look at what they're looking for responsibilities handle data transfer from client AIs media and to external data partners, maintain monitor Data reporting, integrity expert and format quantitative results, perform data mining to extract insights, recognize and communicate meaningful patterns and Data when they put skills comfortable working with large data, set strong analytical skills, the ability to collect, organize and handle. I don't know, like I'm looking at this. I'm like, oh, this.

[00:42:34] You couldn't just call this job data scientist. Strategist. I like it. Yeah. And I think that that's also I mean, not to shadow companies or whatever, but when you put the job patala something like this like strategist's instead of data scientists, all of a sudden they don't have to pay you a data scientist salary. Right. That's kind of like. Yeah, great question. Anything else that I know you should move to.

[00:42:57] Next question Incremented So I have a quick question. Sorry, I was trying to type into my LinkedIn and I couldn't get to the part where you were staying, put it. And so so you said go to the search bar. Right. Not now. Is it under the jobs thing or you click on jobs and then go to the search bar.

[00:43:13] Yeah. So you click on jobs. There's like the little seabag. OK, yeah. You click on jobs and then it's a search for your next job and then search by title skills or company and you could just type it. Typing skills. You might want to do more, maybe keyword search. That's where you put the skills themselves in quotes without that and see what happens. OK, all right. So thank you so much. That's no problem. Where are you located by the way? I forgot I'm located in the Bay Area. OK, yeah.

[00:43:44] You'll probably find all sorts of weird job titles and yeah that's why I like. Wait a minute, I'm not ready for Data science job yet because I still don't know how to model the stuff. I want something that kind of gets me to the Data hole and do Data cleaning exploration Data and I'm comfortable with that. Some kind of looking for those type of jobs that just give me a foot into the door, Data or and then, you know, maybe a few years down the road I become a data scientist or something. But yeah. So the rules are so confusing sometimes and I it's like OK, sometimes the data entry, when I look at them, they want people to do models and stuff it. That doesn't look like data analyst works like a data scientist. And I just go the job description, some of them I just putting everything in there just to get this unique kind of person, you know. So it's yeah. So I was just wondering, I think something here is looking for some people say, oh, go look for product analytics. A business analyst, business analyst for me is more like a project manager. They take the requirement in order to talk to customers and of stuff that's more like business. And I'm also seeing business analytics, but I have all this data job that they're doing. So that's where is the big area. It's just like everything so mixed up.

[00:44:58] Yeah, yeah. So definitely typing the skills look holistically at the entire posting and then based on the description of the job in the posting, then apply for it.

[00:45:07] So when you say the scales right, and if I'm looking for a Data, not a Data side is the scales could be PYT not not a Data science at all, but just any other Odilo it so I don't think looks like Python AIs SQL. Is that what I should be.

[00:45:24] Yeah, definitely typed as well. Just quick, quick heads up like your keep knocking on microphone and it's uh it's oh it's OK. You're making it a difficult thing for people listening on the podcast.

[00:45:36] But like you can put like for example, like Data cleaning. Right. And then maybe Data of I don't know how to spell visualization visualization. And if I type in right now, put Data cleaning Data visualization. And then I've got rules here for reporting visualization and insights. Analyst Power by report developer. I've got it. Storage analyst with um analytics engineer like analytics engineering. A lot of that. Right. Like all these weird like panels that are popping up by analyst, applied research analyst. And I'm looking at like this applied research analyst job description. And this is I'm looking at jobs in Canada. Obviously, that's where I live. And like, it just all looks like the same type of stuff. So that's one way to do it. Put the actual maybe tool itself. But if you want to put the general skilled name, then that will work as well.

[00:46:30] Ok, right. Thank you very much. No problem. So next up on our list, we've got got show back up for another question.

[00:46:39] So my question is, when it comes to especially when you're doing an analysis on your own away from work, so you have that problem of how do I ask the correct question, how do I frame the question, especially when building a model so you know how to grow from it for that domain knowledge? Do I need to have that? Do I need to do a few business courses? And is there anything you can recommend or how do you go about it? For me, I approach problems with the very technical side to it. If it's the numbers good, I'll give them to you that I'm OK with. The problem I mostly have is thinking around it, thinking because most of the times I'm doing something and someone comes. There's an easier way to do this. But most of the knowledge they have is domain knowledge for lack of a better word in terms of business. Are there any business courses you would recommend or not? Is there and how do you face the question?

[00:47:29] Yeah, so, I mean, I don't think there's like a business course that you can take. They'll just teach you how to get domain knowledge in general. But I'd say research and read up about what people are doing in that industry that probably be a good way to do it. I'm a turn this one over to Odali.

[00:47:45] Yeah, I think first try and get really close to any business partners or stakeholders. So if you are currently working in an organization, try and sit in even on their meetings and understand the kinds of problems they're running into, that's probably one of the best ways to get this domain knowledge. I've also recommended this book before, but it helps really technical folks try and think about things on business terms. But Data Signs for Business is a good book that will start to shed light on why you're starting to get these requests for numbers, why you're starting to do these models and how it impacts the rest of your business. So that should be helpful.

[00:48:30] Do you have any context you can put around that idea? Let's just say, for example, you're asking a general question, but I'm sure you're asking because you're working on some specific if you want to talk about the specific case. And then maybe we can give you a set of principles that you can then use to apply to the next problem.

[00:48:47] I know a lot of this is a problem. I've been facing a lot. I think I approach problems from a very technical side. If you want the numbers, I'll give them to you. Them visualizations, I'll do it. But a lot of the times I go to, for example, choosing the correct you into a new field, into a new market so you don't have enough information on customers. So you have to decide who is a good customer to interact with, for example, and giving out loans and was a bad customer. So, for example, you will need to think of how to go through it. I was having this conversation at work. My thought process went to something very complicated when someone just came with a simple solution. They just do a B testing while my head was going through the probabilities. I'll do Baesler like I was going through the whole numbers thinking to things. That is something you just pick up with experience. Or do you think that knowledge.

[00:49:35] It's I think that's what that is. Right. So you just, you know, so much stuff and you just want to apply it. And I think at times it's a very good practice just to pause and just think from the ground up without introducing terminologies that introducing any of these complicated things. Just think about what it is that the actual request is. And I think you get there by having conversations with the stakeholder, by asking questions and and then from the answers to those questions, map of possible solutions. So don't come right out the gate with the solution mindset. Come with a question. My. And they let me figure something out based on the interview with you and then move from there as he tore out his hands up. He probably has some great advice here as well.

[00:50:16] There we go. Yeah, this is common. I mean, I've been working as a business controller and as comptroller and auditor. And, you know, in auditing, we always face new things. I'm auditing in the oil and gas industry. I need to get involved in engineering. I need to get involved in drilling and production and temperature, you name it.

[00:50:38] I think the key is, is when you're the general problem, I find by most people, especially when you meet people that are very skilled, is that they go directly into the complexity of the problems and they basically create so many problems. Nobody wants to listen anymore. And one of the things that I've come to conclude or learned is that starts simple. I mean, you're talking about customers in the bank who are good or bad while the bad customers are simply the ones that don't pay.

[00:51:10] So if you don't know, you start there, that's a good starting point to find out who hasn't paid and then all the customers are going to automatically start coming. The next step is going to be why didn't they pay? Is it the environment that they couldn't didn't have bank access or whatever? And then as you progress, you have to start involving the people that sits with the knowledge so that you will then bring in the marketing people, you will start bringing in the financial staff, you will bring in the different people, and they will then start to give you a picture. Once you have that picture, that's when you are you have been there. You are now in a position where you can actually start applying your skills for me. You cannot solve a problem if you don't understand the problem and to understand your problem, you have to learn.

[00:51:56] And every time I go into a situation, you have to be open and just listen, listen, look. When you get confident, then start approaching with your skill set and you will be challenge guarantee. You know, that's not the way it is that, well, there's nothing wrong about being challenged. That's a good thing. It means that you can now focus more and more. And over time, like in sales, I've always said eliminate the notion you will end with this is not a question. You will always end with the. Yes, because a person who doesn't want to buy, you have to find out what and they eliminate that. And then you go to the next level and then you eliminate that. At the end of the day, there's just. Yes.

[00:52:35] So for me, the whole thing is to start out by asking and keep it simple to start and the complexity will come automatically. You will meet people that will start with the complexity right off the bat. Don't slow them down. You have to control people's expectations, meaning that if they come with a lot of skills, that's a lot of a lot of information. Break it up, slow them down so that you can follow up because you were the one that has to learn. And once you get up to a level where you can follow, then you set the goal and now you can start modeling and really go into the details, because now you said, I absolutely love that.

[00:53:10] I was interviewing David Benjamin earlier this week. You wrote the book Cracking Complexity, and he has a ten step formula for cracking complex problems. And step one is asking really, really good question. So I think it starts questions and then follow up with methodology, ideally. What do you think?

[00:53:26] Yeah, I agree. I go kind of with the five wise. So getting to the root cause of why they're bringing up this project, why these numbers are important to them once you truly have a good understanding.

[00:53:40] And I think that knowing that getting that understanding does take time, unfortunately, is not like when set in meeting or one conversation is not going to make you an expert in it. So knowing that you will be more familiar over time and then you can use all of the other methodologies, you know, and apply them when you're starting to get to the meat of the technical work. So worry you can worry a little bit less about that up front and just try to also flex your ability to push back as well. You'll find that they will make a lot of assumptions that tend to back up their business goals that may not be truly substantiated. So I hate to see you take it with a grain of salt, but you have to. The biggest muscle I had to learn was that we should be thinking we should be pointing out when we are just reinforcing what we think is true and not what we've actually seen evidenced in our particular data set. So being able to flex your ability to say no or to redirect when they bring bring up a certain kind of a project.

[00:54:57] So, yeah, OK. So I don't know. I was going to call on you to see if that answer your question.

[00:55:04] Yes, but OK, first things first. Thank you for the other weekend. That's really helped my other question is, I feel like with this field, I'm consistently learning like it has reached a point where there is no line between my social because I'm consistently taking new classes and courses that I consistently keep meeting new things I don't know. You've been in this field longer than me. How do you go about it? Is there no end there? Just give me two new things and learning new things and taking your classes and going on every weekend. You have a new class to go on, or do you just stick to what you know, perfect that. And that's because every time you go somewhere, you meet someone talking about something brand new you've never heard of. Now you have another class to start taking. When you're done with something else, you start taking another class.

[00:55:48] Yeah, I mean, first of all, yes, you will always be learning as a Data scientist to stay relevant and current in this feel. You're always going to be learning something at some point. Do you need to learn everything that you hear about in the news? No, absolutely not. I don't think you need to just because Data scientist X, Y, Z at ABC Company is learning about one, two, three. Does that mean that I need to learn about that because it's not relevant to what I'm doing in my day to day job. So that's why I'm a huge proponent of just investing time and fundamentals and principles and methodology. And then from there, whenever we need to learn something, you know that you can fall back on a strong foundation to pick anything up going forward. Right. So for me, like I will, I spend more time reviewing statistics and classical machine learning and principles and processes of how to solve a question, how to ask a good question than I do learning about deep learning stuff. Um, that's just to my perspective. I don't think you need to learn everything. Do I always learn? Yeah. I spend a huge amount of my time learning. Literally every waking minute I would say of my day is spent learning something like you look around my desk, I've got no fewer than 12 books on my desk right now. I mean, the phone books and books for people that I'm interviewing, which is cool. But yeah, I think that's just part of being a data scientist is continuous learning and improvement. And I think that's what makes the job so amazing. And fortunately for me, like on the job, like I get ample opportunities to study on the job. Um, you know, like my framing is I wouldn't learn this stuff outside of work. I'm learning this stuff to add value to this company. Therefore, this is worked there for it to research. That's my perspective. I see Natasha as needed. So let's hear from Natasha and let's hear from you on this.

[00:57:31] Oh, I was just going to say I can vouch for what Harp is saying because we worked together. He was. Yeah. Constantly learning on the job. That's that's the way to be.

[00:57:41] Yeah. And back then I was taking actuarial exams. So, yeah, they literally were paying me to not only study for the exams, but they paid me to pass the exams as well, which was nice by having to study on the job. This is key, but don't feel like you have to learn everything. I think that's that's really, really important. I love to hear from Ideal and then whoever else wants to jump in on this man, let's hear from you guys as well.

[00:58:04] Yeah, I double down on not feeling overwhelmed, but instead trying to see as you have the opportunity to learn all of these things, sometimes it's hard to tell what is important for your job or for your role and what is like a the hot new language model. But I think trying to take that pressure off of yourself that you have to know everything that's not the case. And no, no, when does I think when I started to get into the mindset that I will always be behind in something and it's just an opportunity and recognizing that so many people, specifically in Data science and machine learning, like compared to several other fields, are so open and willing to talk and excited to tell people about the things that they know. So I've gone to conferences and been like, I don't understand what this concept is. And someone who was like, oh, I wrote some of the foundational code will go through the basics with me, like we are lucky that we're in an industry like that. So recognizing that as well, that there is probably less looking down like you should know this thing and people are more likely just willing to help you either understand the concept or give you resources, book advice and recommendations.

[00:59:26] And another thing I found helpful to manage everything is just something as simple as this. It's a weekly to do list. Right. And you can see here, like these are all things I'm doing Monday, Tuesday, so on and so forth. Here's like daily things that I'm doing. Here's other stuff that I want to do during the week. And, you know, I'm not going to check every single box every day, but at least I know where I'm headed, what the week is going to be like. And it might be helpful for you to manage your life in terms of seasons of intellectual curiosity. Right. So it might be that, you know, over the next seven weeks, I'm going to focus entirely on understanding customer churn and understanding what that is all about. Right. And if you do that times. Six or eight, what, seven times? Fifty six, right? Yes, so that's a year so you can learn eight things in a year that way by having seasons of intellectual curiosity. And I find that to be very manageable and helps because, dude, if you try to do like day one day, I want to do this next time and do that. And you don't really make progress. I don't know if that makes sense, if that's helpful at all or not.

[01:00:31] Yes, it's very helpful. At least I know the consistent learning is not only me, I'm not looking down and not stepping back to look at the bigger picture at everyone's doing it.

[01:00:40] See, Tau has its hand. That was also enhanced from the last time.

[01:00:44] You know, it's not from last time. I, I just want to comment on this because, you know, I've been in the working environment for too long.

[01:00:54] I like to retire yesterday, clean up because of the hard work, but more the experience. But for me, the key is that, I mean, it's impossible to learn everything and that is just not possible. Don't waste your time. Don't waste your time trying, and don't waste your time taking a lot of courses, because at the end of the day, if you don't use what you're learning in that course, I give it three months, four months, and there's got it has no value. The only value you will get is actually, like I said that is that if you take a topic, learn the basics that will give you the basics. So later on, if you happen to get into a situation where you need to learn, you remember there was an article on that. There was I remember that situation that that's when you if you need it, then go and take the course, especially if you're going to pay for it. Other than that, to me, it's really completely which is better to work on your skill set and improve what you are actually doing right now in your job. Take your energy and spend it wisely because there is only so much and you need to have a life outside. Work isn't everything. Lesson learned the hard way when you run 12, 14 hour days for five years straight and seven days a week, there is a wall and it's really hard and you definitely don't want to head now.

[01:02:17] Some people enjoy reading books. I am not one of those. I like learning by experience and hands on, etc.. If you are the kind of person that enjoy reading books, do it. I mean, they let you have Friday nights are night with your books. If you don't, that is better to talk to people like in conferences. Join these type of settings where you can get introduced to new things and what's going on. And this podcast. Perfect. I mean, I have absolutely no idea about that science and all of that python and all that stuff. But I'm here because it's a keen interest to me. I'm not going to become at that interest. It's too late for me to like I said, I don't want to go back to school. I don't want to learn all of this. But I am gaining a lot of knowledge from the people here on this podcast and the group that I can then utilize when I'm talking to the professionals, which will come into my life in the next few months in regards to my project. And I need to be able to understand what they're talking about. That's what I'm learning to that.

[01:03:20] Thank you very much towards super happy to have you here. I know I've learned a lot from you and I know the audience as well. So thank you very much for that. And I would say just before investing too much time in learning anything new right up front right now, spend some time on learning how to how to actually learn. That's something that we have never been taught. You know, I've gone through years of school and nobody's ever taught me how to actually learn. And I think that's kind of a disservice. And once you learn how to learn effectively and efficiently, then things just become a little bit easier.

[01:03:50] So you can go right now and go to Coursera, get the course, learning how to learn, taught by Dr. Barbara Oakley, completely free. It'll take you no more than three to four hours to get through. She also has a book that accompanies the course called To Mind for Numbers, which is all about learning how to learn. I will be releasing a podcast episode with her later this year. Um, she's a lovely, lovely lady, learning how to learn. And if you're in my community, I've just shared this resource, pragmatic thinking and learning. This book is I'm telling you, man, this book is frickin amazing. Um, so another book I recommend on top of that is just Jim Quix Limitless. That's another good book. So Learning How to learn. Very, very good course on Coursera and very good skill for you to have as well. Thank you. So Quinten actually has a question I think is related to what we're talking about right now. So Quinten, go ahead and ask a question. Then after Quinten will go to Davran the ash and then we'll call it a day. So if you have a question, now's the time to let me know and put it into the queue. Otherwise, after Quinten, we'll go to Dan Rather than the ash.

[01:04:52] Yeah. Hi, everyone. The other question is in the same line, basically, do we have like some kind of golden resource where we find, like all the relevant materials that we can learn the new tools that are coming up that are that are meaningful like. The thing is, I have the same the same thing where I'm sometimes dispersed, scattered all over the place and trying to run everything. And at the end of the day, I'm getting frustrated because I'm not learning, actually, because actual mention is the practical and I'm forgetting and whatever. And then it's a vicious cycle. But at the same time, I completely agree with what you're saying.

[01:05:26] I've decided to focus on something for seven weeks, like really internalize all the concepts of basics, and then you can progress into the complexity of it and then you can actually get something. And I think like when you have the work already, you can just basically focus on the numbing knowledge of your actual work, that sometimes maybe the tools that you're going to be learning in your work environments are not that relevant compared to the mass of the market. So you like you know, you're learning something, but maybe it's not that relevant compared to what's actually going on. So do we have some kind of resource or methodology to know, OK, this is the tool we should be learning? Maybe this is another tool we should be learning and maybe that's all we should be learning. We should say to the company we are working in that we should be using this tool instead of the other ones, kind of like knowing how to keep up with the right environments, because when you're learning a tool like we know now, Docker is kind of very important. But it has taken some time to get to that point like I did for the very beginning. People were probably in the mindset of, OK, should I get into this or should I not or whatever, and then I'm going to invest some time in instead of wasted that maybe you should have spent the time into something else more valuable. So it's always like a trade off that is complicated to make.

[01:06:41] I think when it comes to keeping up on what new technologies are out there, going to conferences I think is probably a good way to do that. So like PyCon Pi Data conference or attending webinars, like the ones that comment I host quite often, those are useful as well. I think there's they have these panel discussions as well with a bunch of people talking about Mellops, stuff like that. So that I think is a great way, but I don't know if that fully answers your question. I'm still kind of digesting the questions, so I'm going to flip this one over tiredly.

[01:07:13] I would say I have not been able to find any kind of golden source that has everything, because I think part of it is that science is kind of hot and because so everyone wants to create content about it, I think is the biggest problem. But if you take this focused approach, you can kind of create your own digital blinders that you have maybe your areas of focus as different folders of browser bookmarks and as you come across things that you want to learn in three months, you just save it as this. I'll hold off and not get myself overwhelmed. And then so that's kind of been my approach because it is so easy. If I had just one learning folder, would have seven hundred links in it. So if you can try and create those ways to flow through the different topics without feeling overwhelmed, it might just have to be organization because it is. There's great stuff on Medium as well as Coursera. You've done kind of everywhere.

[01:08:19] Natural definition I think it is. Yeah. I guess you always have to be like on the on the spot like every day looking at every sources and keeping up to date, but not being overwhelmed. Yeah, fine.

[01:08:33] For me personally do it. I don't really focus that much on all the new trends and technologies and stuff like that. I just kind of, I kind of like to meet the basics. The fundamentals, the foundations of things are super important. Um, but I have colleagues like, for example, the Data architect I work with. He's always up on new trends and I'll rely on him as a source of information for what's new. And also just, you know, things like this where people talk about what they're working on or happy hours, office hours or people talk about what they're working on. And it just gives me new things to explore. So LinkedIn is a great choice as well, I think just following what people are talking about. So hopefully that was helpful. Hopefully get the answers you're looking for completely. Yeah, no problem. Let's go to Davran and then after Davran we will go to Yosh. Oh Natasha, wait. Do you have an input here. You're muted. We can't hear you.

[01:09:24] Tasha, can you hear me now. Yes. OK, and you guys sort of Data cam.

[01:09:28] Oh it's like it's like boot camp platform.

[01:09:30] Right. So it's an intuitive like the interface is like you don't have to download any of the Data tools. So it's like an interactive way module type of structure. So there on this source you can see all the latest technologies. The ones I mentioned are SQL, Python, Tablo powered by scale. I shall get and you can pick also which topic you want to. Or get better at so for example, you have probably in statistics, importing and cleaning Data, applied finance programing, data manipulation, data engineering, so you can pick the tool and then you can pick the associated topic or you can just pick the tool and then it'll give you a variety of courses. So it is a theory, of course, is very important to learn and then be being able to apply that. So I think this Data camp is great. Oh, also, you can do case studies. There's a bunch of stuff on like covid related studies.

[01:10:47] Yeah, I highly recommend this. And then you can also earn certifications and share that on LinkedIn. But I know someone earlier was just mentioning like, oh like everything is like scattered. But I think this is a good summary of all the different technologies and topics you can learn about myself.

[01:11:06] Also looks like it's reasonably priced as well. So that's a great yeah.

[01:11:10] And I think it's like 30 bucks a month. So it's I think that's reasonably priced.

[01:11:18] Thank you for that. So bunch of awesome resources for you there, Quentin. Let's go ahead and move on to Davran.

[01:11:24] I, I didn't mean Harpreet. I would like to first thank you for all you guys do, giving us so much to Data science community. My question is regarding I listen to a podcast about the interview challenge challenges. That's how hard it could be. Right. But my question here specifically about the coding challenge, the python coding challenge, like when we first learned Python, right. How HelloWallet and show us about Data structures, lists and dictionaries. We expect it to us getting this kind of basic questions or more like it sounds like a building help helper functions and python building, how the Data science process goes from idea all the way to the production up.

[01:12:04] So I guess to just make sure I understand the question somewhere rephrases back. So the question is for coding challenges. What types of what types of coding interviews, what types of topics come up frequently? Is that question?

[01:12:17] Yeah, kind of like the like, for example, code each other. What kind of is it like more heart type of coding will get or more like very basic questions we're getting like what is less what is like dictionary.

[01:12:28] Yeah. So the questions I've seen on coding challenges like the coding screen interviews, they're definitely not easy ones that are kind of puzzles. And the problem solving question is essentially right. So you'll be given a problem statement. And so you need to find or create a python function that satisfies the conditions and passes all the test case for that particular problem that they have their toughman. They're not easy and they're beyond like that fellow world level one website I found super helpful to just kind of get you get you prepped up for that python principles. I'll go ahead and I'll find the link and I'll share the link here while I tell you is giving her a response.

[01:13:11] Yeah, I would say Hacker Renkin, the code in that medium to hard level is pretty much where you're going to be looking. So despite the fact that they may not be as specific problems, they may not be design specific problems, you'll see very much algorithms and kind of data structures. But and being able to get, I would say, comfortable with at least being able to attempt maybe some of the hard problems and being able to complete the medium ones. That's the vast majority of the kind of coding challenges I've had. You may have some like take homes or they ask you to maybe write some secret code or conduct a very basic analysis. But yeah, those are harder level python questions. That's what you'll most likely run into.

[01:14:05] Yeah. That are challenging, man. So don't get your confidence beat up if you're going through those and you're like, oh my God, this is super difficult. It's meant to stretch you and you get better. So the website you linked here, Python Principles, challenges, this will be a good way to to kind of build confidence in yourself and kind of understand how to solve these type of challenges. And then after that move up to like rank and some of the more difficult, um, levels they have their ushers talking about pi for e dot com. I've never heard that. So I'll definitely look into that. Yeah, hacker is definitely a good source as well, but I get up to. Yeah, yeah. I'm sorry to cut you off, just finish yours and say I think the most challenging part is taking the Concord word problem that they've given you and then trying to translate that word problem into a programing problem. That's going to be the biggest leap, because once you've been able to understand this word problem as a coding problem, like the.

[01:15:06] Functions aren't really that difficult, which is interesting, but, yeah, sorry going on, is it is it all of the is it mostly will be like your take home or they will be like next to you watching you with you think it could be a mix of both.

[01:15:21] So typically there'll be a coding screen that comes up and you'll have maybe like an barcoding screen. I mean just like they'll send you a link to some platform and then you have a set time limit to go through maybe four to five problems. And that problem could be a mix of python SQL machine learning stats depends entirely on the role. And then once you get past that, then there's still more to do, which could include a whiteboarding challenge.

[01:15:46] Ok, thank you, guys. Thank you very much.

[01:15:48] All Yeah. So let's hear from Aisha than Quentin on this topic.

[01:15:51] I wanted to mention the site of LinkedIn. It's called For It by Chuck Sevres. I think that's the name of the instructor, but it's just for introduction to Python. You've never worked with Python. It doesn't have the data science library is like the not by the Sibai. It doesn't have all that. But it'll give you open like the first step of how to step into it before you walk into the python libraries. And for the code. A lot of the times, especially with the libraries, the Data science libraries, you can automate very many things. So you do not have to worry about that. That's my opinion, because a lot of this time, a lot of the things you just call it a function, then just carries it out for you. So you do not have to worry about the metrics. Some of them are very straightforward, it looks like. Thank you.

[01:16:35] Thank you very much. Yeah, it looks like it has an accompanying book that goes with it. And the website seems like all the stuff here is free. So that's awesome as well. Quinten.

[01:16:44] Yes. Yes. The question is, I discovered recently one of the results like occurring.

[01:16:52] Yeah, I think I think it's really good.

[01:16:54] Like I was trying to do challenges. Some of them were quite difficult, but I felt that they were kind of software engineer related, like they were not about like it was about python skills, wasn't about Data sense, what python skills. So regarding those kind of challenges, plus OPIS or object oriented programing, like how much of that is kind of required since you guys have kind of more experience in that Assange job?

[01:17:25] Or is it or is it just kind of a plus like a bonus for the employer to say, OK, this guy has a lot of python skills, so maybe he can move into something else more specific? And not only that assigns jobs, but how much is that related to that?

[01:17:38] As I think he's pretty so object oriented programing is pretty important to do science roles. So definitely know how to do that ideally.

[01:17:46] Yeah, I would say despite the fact that it feels a lot more software engineering. So I would say that that's the norm, especially for interviews. You may notice that you're doing things or using frameworks or tools to actually do it once you get into the role. But it's kind of the norm to expect that that is the what you're getting tested in interview versus building a model or being able to use any kind of like Data science libraries. So, yeah, it's it may not totally match up with when you do start doing the work, but having that expectation that you should still have a good coding and programing background and thinking about it as well, like you tested on the problem solving mindset of the thinking process, not just something the product side is OK, I think you're right on just noticing the time.

[01:18:42] We'll wrap it up with one more question from Yosh hour and a half by super quick, guys. Great questions.

[01:18:47] Josh, go ahead and we'll finish up with your question as I'm currently pursuing bachelors in Applied Statistics Analytics and the second year of this, bachelors will end in June or July.

[01:18:59] And I left only one year after this. My question is, should I go for masters after bachelors or should I go for get some work experience of MOOCs both here, because both of them are pros and cons. Some people say that you should go for Masters. What really does I afford it? I stores for a company. They don't ask for masters. And some people say you don't have to do a masters after you get into a job.

[01:19:32] Yeah, this question I would say you need to answer for yourself. Right? Like, why do you want a masters? Because if you want a masters, forget what everybody else says and just go for it. Right. If you want to go and study more and study deeper into a particular topic, then you have to go for it, study, get their masters degree, get better and understand something at a more deeper level for sure. But if you just want to get a job and start making money, then is getting a masters aligned with that goal. So what is your goal in life? Right, to find that for yourself? If your goal is to have a deeper, more intimate understanding of a particular field which you believe that you can only gain through. For the graduate education, then go down that path, if you if your goal is to. I need to get paid man. I need to make money and yeah, I want to get paid, then go get a job. Right. I think as simple as that, but ideally go for it.

[01:20:18] Yeah. I agree with you. It's about short and long term goals. So if you are in the stage where you really want to start money making money, start getting work experience, I would start looking for jobs.

[01:20:33] But if you are ready to kind of go deeper than maybe a master's degree is something that actually I think according to my I think that is not, you know, like another for lack of chemistry or physics. It's a tool which implies in everything you can apply Data things in sociology, chemistry, chemistry. You can apply everything there. And all my bachelors, they are completely full of all skills. They are teaching me machine learning are fun. So before I board in the meeting, I'm also doing quantum computing and quantum computing. Have I think have I, you know, commentary every Data since, you know, cryptocurrency. So value mining, the cryptocurrency, you know, the quantum cryptocurrency also and then Data science scheme for mining the cryptocurrency for and for the AIs for my studies for if you want to do mining quantum cryptocurrency for now they are asking for master's degree at least. So that's why I raised this question.

[01:21:45] It sounds like you already have the answer to a question. Yeah. So is it in line to goals like that? That's what it comes down to, right. So it sounds to me you talked about you're interested in like cryptocurrency, quantum computing and in order for you to go for an interest in research field. OK, well, if you're interested in research, then, yes, go to grad school because that's where you go. You learn how to research. I don't think they really teach you how to research an undergraduate program, but it comes down to what do you want and what aligns well with your goals. That's ultimately what it comes down to. It's not an easy question that we can say, yes, you, based on the twelve minutes that I've talked to you, should go and do a master's degree. I, I can't I can't make that decision for you. You got to you got to really think about what you want. What are you trying to go and if one path or the other is going to get you there.

[01:22:39] So yeah that's that's what I also asked this question with another people like my and I asked my elder brother. So he said that you should wait at least two to three hour work experience. After that you go for master's because at least you really get that how to mine. Did it all legalized? Yeah.

[01:23:01] So I think I should research for this, then figure out what works best for you and then go down that path. I see Quinten and get your hands up, guys. We're going to wrap it up here with that. So I appreciate you guys wanting to chime in, but we're going to go ahead and wrap up today's session. Thank you guys so much for taking time out of your time. But yeah, no worries, man. Thank you for taking time out your schedule to be here, guys. We'll be back again next week, same time. But before we go, I want to remind you guys that the Data Community Content Creators Award is open for your votes and your nominations. So we can't make this happen unless you guys vote. So here's a link right here to go and vote for your favorite content creators among various different fields hosted by Kate Strachan and myself on LinkedIn. Completely live and totally powered by you guys. Like, literally, we can't do this event unless you guys are out there voting and helping make this happen. So please take time out your schedule to nominate your favorite content creators. And let's make this big man I'm excited for. This is going to be huge red carpet event. Madness would be awesome. Also, don't forget, tune into podcast @TheArtistsOfDataScience or at least an episode on Friday with Dr. Sutton a she had Gellatly just she studies comedians, which is awesome. Don't forget also Friday after our session joined me on that as well. This thing will be up Thursday, as usual. Guys, thank you so much for hanging out. Appreciate you guys being here. Take care. Remember, you got one life on this planet when I tried to do some big cheers.