 Hey everybody. This is Monica Wahee going live with our data science chat today on January 22nd. I'm just going live about five minutes early to give everybody a chance to connect. I tried to publicize this one a little bit more than the previous ones because I'm like learning how to do this better. You can watch me on a learning curve. Those of you who are sticking with me. So let's hope that more people join the chat. But of course I joined, I set up five minutes early on purpose to give everybody a chance to join. So I'll just talk about today. My goal was to get a discussion going about how to prepare for job interviews in data science. And you might be like, well, why are they any different than any other job interviews? Well, in some ways they're the same, but in some ways they're different. So and of course every data science position is different in every organization is different and so every interview is different. But there's certain things that turn into a pattern that you start seeing with these interviews. So I thought, well, let's see if we can get some advice going. Well, thank you. I'm seeing some people in the chat. Moriah, is that how you pronounce your name? I'm so glad to meet you. I'm sorry I'm reading, I could show it. So Moriah says good morning, my so nice to have you set this up while I'm getting ready to head out for a hike. Sorry about that. Actually, you can bring me along on your hike. I don't know if that makes a very good hike. But and then also Tanya, thank you for showing up. I appreciate it. So that's again, so I'm a data scientist. My name is Monica. And you know, I'll be honest, I have a Moriah, Moriah. Okay, I'll say Moriah now. And oh, if you thank me, you get to be highlighted. Ovi, thank you Monica for doing this. I hope you get something out of it. I hope it's good for you. So what I was just kind of saying to fill up time while people are joining is that if you're interviewing in data science, you'll find, you know, each job interview is like for each position might be structured differently, you might be talking with one person, you might be talking with multiple people, you might be in a group interview. But what you'll find is you end up talking a lot like you end up in a lot of interviews and sort of like, you sort of feel like you're constantly repeating yourself in a way like after a few interviews. I admit it was always really easy for me to get a job because people just didn't want to pay me anything. And they just, you know, you know, they just thought I'd work for nothing. So they would hire me. It was not easy for me to get a job and negotiate a salary that was decent. But it was always easy for me to get a job. So I never did a lot of interviews. Oh, thank you for joining. Happy new year to you too. Thank you. Yeah, so I thought that and I invited people that I know don't they aren't looking for jobs, but they have jobs. So if you have a job, you got it somehow, right? And I remember back when I the last time I went on a like job interview, like a job search was in 2011. So, you know, you might think, oh, what you have to say is dated. But while I was a little worried about that, but I actually wrote this blog post on it. And hopefully Joe will show up. He's my friend from LinkedIn. And Joe actually recently changed jobs. And he told me what I said on the blog post was dead on. That's what he said. So I'm glad that my, my information is still up to date. But I was pretty sure it was because I get a lot of customers who my customers often are in college. And so they graduate and then they need to get a job, right? And so also I have customers who are already graduated. But then I was thinking, Oh, you know, I'm aiming this at people looking for jobs in data science, who are like young, like they've never had a job before. But then I realized I just went on a job search in 2011. And I was not young anymore. So this is really just about the interviewing part of your data science job searches. Let me see here. Oh, I'm kidding. All right, so now it's noon in Eastern time. And so I'm going to open up the officially start the live stream. So hello, everybody. Thank you for showing up. The purpose of this live stream was to discuss advice that I have and any of you have for preparing for job interviews in data science. And like I was saying, when I was filling up some time before I officially opened the live stream was that maybe this advice counts for all different kinds of interviews. But in data science, we end up being often in interviews where there's there's either one person or a group of people and we're being asked a lot of questions. And it's not like we have one interview and then we're hired or not hired. It's like we often have to go through rounds of interviews. So it's like you have to get good at being interviewed for data science position. So how do you do that? So I started I thought I'd start the conversation by sharing on a blog post what I do with my learners, my customers, when they have to get a job and I have to prepare them like I'm the person preparing them. So let me share my screen here. Let's see. I know how to do that. So actually let me yeah let me share like this. So this was this is the blog post that and I guess I could just share it in a chat too. Hopefully you'll get this. Hi, Rizul. Thank you for coming. All right. So here is my blog post and you'll see that I sort of boiled down to three tips. But tip number one is kind of a big tip. So maybe we should talk about that one for a little while. And that is to make sure you have at least one project to talk about authentically. This is a situation where if you're new to data science this is can be kind of a big deal. Let's see. Yeah, this can be kind of a problem because oh hi, Mohammed. Welcome. Welcome. So if you're like some of my colleagues or like that Joe guy who's going to join, this is not a problem for you because Joe and I we've had a lot of jobs. We've had a lot of projects and we can talk authentically about them. In fact, if you take my LinkedIn learning courses or you work with me at all, I talk authentically about all kinds of stuff I've done. Well, hello. More from the chat. Oh, thank you, Musa. Musa says my field is not in data science but this is absolutely great. What are you doing? Well, I hope you can use the advice that I'm giving you because in data science we're always using projects and then talking about the project. Oh, Joe's here. Joe's here. Great. I was talking about you, Musa. So Joe in the chat here, if you're on LinkedIn, if you're on YouTube you might not be able to see Joe in the chat, but Joe recently landed a good job like it was the kind of job he wanted and he's the one who sort of validated my blog post here. Let me switch how I'm displaying it here over here. That way you can see it more. I've got to get good with this re-stream. So this first, hi there. Hi. Another hello from the chat. So this first tip, make sure you have at least one project to talk about authentically. Joe and I don't have any problem with that, but if you're new to data science and you haven't had a job yet. So I have an intern and she used to be a musician. I mean she still is a musician, but so she's kind of in that situation, even though she's had jobs before, she's just never had a day-sized job. But she's doing the right thing, I guess, because she's my intern, she was to me. And she's actually kind of spreading herself then. She's working on a lot of different data science projects. And in the beginning she was, I think she was a little misguided because she was just kind of making dashboards and things, but you got to start somewhere. So I don't want to dissuade her. Now she's more into completing projects. Like where you start with a question and you answer it. And in science, you do it with a paper. But you can, in data science, you can kind of do it with a solution, you know, like here is our problem and here is our solution. But you have to be able to write it up and communicate it. And so if you think about this, make sure you have at least one project to talk about authentically. If you think about a project where you go from, I don't know what to do or whatever, to the entire solution and you write it up, that's like an experience. Like you change when that happens. And I'll give you the example. So you change and you become different. Even if you didn't have any interest in the subject before, you didn't know much about the subject, especially if you're in a group of people who know about the subject, you might be like, okay, whatever, I'll do your statistics at the beginning. And at the end, you're like, oh my gosh, blood transfusion or whatever it was about, you get really wrapped up in the subject. And then you authentically just start saying blood transfusion, you know, at the interview, right? And so, so, oh, hi, Rothenam, I'm so glad that you showed up because it must be the middle of the night for you. Hopefully you're enjoying a beautiful night over there where you are. Thank you for joining. So let's see here. So how do you have one project to talk about authentically if you're new to this? Well, I kind of gave an example on the blog. Let's see here if I can find, yeah, like here, see this video? This video is based on a portfolio, I call it portfolio project I did. Why did I do the portfolio project? Well, you know, I'm already a data scientist. Well, I'm a health data scientist. And locally, I live in Massachusetts in the US and we pay taxes. And some of it goes to set up casinos. I know that doesn't make sense, but they have to be regulated. And some of our taxes goes to regulate the casinos. And one of the regulations is they have to hire local businesses to do stuff. So my friend who's a local data science fintech person and me, you know, so she's, she does like, you know, she knows about taxes and accounting. And I know about health. Okay. So how are we going to make the casino think that they should hire us? Right? So that's why I did this portfolio project. And also, I'll be honest, I did it because I wanted to have an example, an authentic example to talk to you about. I wanted to be able to be interviewed by you guys. So, so if we go over here, you'll see this is the video from it. But if you go over here to the blog post, let me see if we stream. So this is the portfolio project. So, so you see how I actually posted this. Where am I? See, I actually posted this online. And that's one of the things I say in my blog posts is you got to post this stuff online. And you got to use like, I made a video because I thought, imagine, you know, I write to them and I say, hey, my business and my colleagues business, we did this analysis, blah, blah, blah. Do you think they're going to read it? No. But maybe they'll click on this video, right? Maybe, and then I can show them like I'm really good at communicating. I'm going to make PowerPoint slides. Look at this. I'm going to make data visualization sounds like you guys, right? Like data scientists, right? So I can showcase that in a video on YouTube and the whole world can find it like, if a casino from Minnesota where I used to live when I grew up, if they come to me and say, oh, Monica, work for us. Do you think I'm going to say no? You know, so I put it out there, make it nice and professional. Maybe somebody will find the video. But then, in addition, you know, the blog post is fun to read. If you read the blog post, and you'll see that a lot of the blog posts is really about the subject manner. That kind of gets back to what I was saying is, if you go to my, like my data science toolkit blog post, which I linked to on there, which is sort of how to get started in data science, I see, let me actually, so Rizul says, I want to learn data science. Can you help me? The answer is yes, I probably can because I've helped a lot of people. But what, where I would tell you to start is with, let me look it up here. Let's see. Is I post of this blog post? I even made a video. I think maybe the video is a little bit better. Let's see here. I'm going to, I'm going to go over there. I'm going to change this. Let's see if you can see it. This healthcare data science newbie do it yourself starter kit. With that, what I do is I'll send you this link on the chat. So what am I saying here? I'll sum it up. The problem is, the video is a little more generic. The blog post is more focused on health data science, but in the video is more like just data science, but you basically need to start with a topic. Like if you're, if you want to do US healthcare, I strongly advise you to watch my playlist of these lectures on the US healthcare system, which is depressing. I'm warning you. But anyway, why? Because you need to have subject matter expertise in order to wow anybody or convince anybody in the field, which is why I did this casino thing because I love being a casino patron, like a customer. But I never thought of like, what do casinos think of business wise? Like if I'm giving, they give me their data and I analyze it like, what am I going to tell them? What am I even thinking about? And that's what that project did for me. So I can speak authentically when I meet with them. I might not be like the biggest expert, but they're the biggest expert. I'm expecting them to. I just want to get in there and make it so that they know that we can communicate. They can teach me about their business and then I can teach them ways that I can do their data and answer their questions and maybe improve their business. So I strongly suggest a result that you read that blog post and you sort of pick an area you want to study. And of course, it's easier. Let me share the other one. It's easier if you are like, Joe and me, where you've had a few jobs and you've been around a while because you kind of know a topic, right? Oh yeah, this is the one I want. So this was, oh actually, I linked to it on here. So in order to, so getting back to tip number one is you need to make sure you have one project that you can talk about authentically. So you don't want to fear like a data science poser. So what's a data science poser? It's somebody who's taken like a whole bunch of courses like in Python or R or whatever, but they haven't really done a project where they're answering a research question. And they haven't really worked in a group. They haven't really done things like that. That's why it's better to do, to answer a bunch of smaller questions. Like if you look at the Casino analysis I do, it's very simple. There's no regression or AI or anything. It's just some ratios and a time series slide. Things probably nobody ever did with the data. Like I went to run a data warehouse at the Army. I did that from 2008 to 2011. And the first thing I did there is I said, it was focused on rates of injury. I said, well, what are the rates of injury? I'm like, what are the rates of army? Like how many people are in the army each month? Nobody really knew a lot of the basic numbers because it's not that sexy. I guess, you know, a cart, you know, diagram is sexy, but a rate is not that sexy. So we didn't know anything about it. So I did a lot of descriptive statistics just about our data, because think about it, a data warehouse, it's never changing like 2000s over, you know, so whatever you could do those descriptive statistics. But anyway, so it's, so this is just a sum of the first tip. You want to do, I guess I call portfolio projects. And I made that blog post and what happens is some people go there and they sort of get what I'm talking about. And some people go there and say, Monique, I can't do that. You can do that, but I can't do that. I can't, I wouldn't have been able to, like if I had picked casinos, I wouldn't have been able to find the data or that you found, which is public data. Or even if I had found that data, this is kind of what I get a lot is I wouldn't have known what to do with it. Now I tell people to go on LinkedIn learning and take my data curation course and my study design one and two courses. And in my mind, that solves that problem. If you know data curation and studies I wanted to, you should be able to take those, you should be able to find data and answer questions. But the reality is, I already know how to do that, like it's maybe not that easy, right? So if you're struggling there, then that's when you should contact me and weaken me for a free consultation. I can throw out some ideas for you or maybe you can become my customer if you need help. It's up to you, but I'll give you a free consultation and try to find you some data maybe or help you hone up a research question. But I would really strongly, if before I do a free consultation with you, I would strongly recommend you taking the data curation course and the study design one and two course because that's, I'm going to be talking a lot about those terminology. And just to let you know, study design one and two on my LinkedIn learning courses is based on epidemiologic study design. And so it's focused on causes, which is great because you're always trying to figure out what caused what. And so this nice set of tools for applying to big data sets if you want to answer a research question. Okay, so I don't see any questions for chat. I keep looking at the chat. If you have any, let me know. Okay. And then the second tip was manage the interview from the bottom. And what I mean by that is what I noticed when I was interviewing is somebody asked me a question and I'd start to answer it. And then I realized I didn't really prepare an answer and I didn't really remember what they said after a while. So I realized I should really repeat the question back, but it gives you a chance to reframe the question. Oh, Moses says, please tell me, can I rewatch this video on LinkedIn? Yes, you can rewatch this video on LinkedIn at the same event page where this event was where you're watching this now. If you keep the link, you can also go on YouTube. I'm streaming this to YouTube. And also you might want to prefer to watch it on YouTube because I try to put in time links so you can skip to the topics because let me go all over the place. But either way, and also I'm really trying to build up my YouTube channel. So if you are watching this on LinkedIn and you want to subscribe to my YouTube, I'd really appreciate it. All right. So I managed the interview from the bottom. What I was saying is people would ask me questions and I'd just start talking, you know, they'd say, well, give us an example of when you did data integration. And I'd be like, oh, yeah, well, we have this data over here and that data over here. And I had to take the census and blah, blah, blah. I'm pretty sure I'm just talking, right? Oh, great. Welcome from France. This is from France. Yay. Okay, we're truly international. All right. And so I just kind of go on and on. And I realized I had, I knew what I was going to say, but not really, like I didn't and then I'd forget what the question was. So let's say we have that question again, you know, can you give me an example of when you did data integration? What is data integration anyway? Like that's a really generic term. So what would be perfect is if I had had some prepared answers of things I like to talk about, like for example, I worked at the data warehouse. So if I had a prepared answer, I'd say, oh, well, when you say data integration, I'm going to, I'm going to see that as like warehousing, because I worked at a data warehouse. And I, so you see, that's where I reframe the question. You know, data integration, I said, oh, you mean data warehousing, right? Where I integrated data? I mean, they can stop you if you're wrong. But, but I give a few, I give a few examples on here, just like of, of like sort of role-playing. But the problem with these examples, I mean, they just kind of show you, but that's like showing you, you know, a recipe and then the, the completed food. It doesn't really show you how I formulated these answers or how I went from the interview question to my answer. Like I can't really show you because it's a blog post and it's not that easy, right? Like you can't teach everything in a blog post. So what I say is that tip number three is you need to practice making succinct answers to questions. Like I was saying about the data warehouse, you know, like if I want to talk about working at the data warehouse, I can practice answers that and then I can use that, that reframing question tactic to try to get my talking points in. And actually to be honest with you, I, I have a friend who's a physician and he was in a country that was going to host a world summit and he was in the downtown of the country and he was running a blood bank and the news had to interview him about how prepared his blood bank was if anything happened. You know, if anybody knows anything about blood banks, they're always prepared for like everything. So I was like, well, you're prepared for everything so you don't have to worry about anything. He goes, well, obviously I'm scared to be on the news. I'm a physician. I don't, I've never been on the news. And I was like, oh, what happened? And he explained to me that there, his organization prepared him by giving him talking points, like stuff they wanted the community to know about the blood bank, like how much donations they got and how they, how they transported blood and emergencies, you know, just stuff that they wanted the community to know. And then they just told them it doesn't matter what they asked, just answer one of these questions, right? Well, you know, if the reporter asks, aren't you afraid that you'll spill some blood? You know, he could just look up for the answers as well, you know, XYZ blood foundation transports over 200,000 units of blood per year or whatever. You don't even need to just look up something. It's not as easy in a data science job interview. You can't just, you know, look up one of your talking points and just, you know, start saying something else that's totally unrelated to who feel like, I mean, you go ahead. But it's better to sort of reframe the question and connect it with what you're saying. Like I gave the example of the person says data integration for me. And I go, oh, well, I'll assume you mean integration in a warehouse, right? Go on and answer the question. So now I'm older and I've had these experiences a lot. So it's not that hard for me to prepare these answers on my own and my home. But it's hard. But sometimes for some people it's hard to. I'm also like a talkative person. So I really talk about preparing answers and then getting somebody who really cares about you. I mentioned in here to practice interviewing with you and to really role play. Like what you would do is get actual job descriptions that you're applying for and show it to this person. This person should be somebody who actually cares about you getting a job. Like your teacher or a loved one. But they don't have to know about data science. They just have to know that they just have to care about you, right? And be willing to help. And so you show them these position descriptions, you know, some complicated language. Just tell them don't worry about it. Then you want to go online and look up questions you think you're going to get. That's actually pretty easy. You know, often my customers come to me and they're like, Monika, help me. And I'm like, did you look up any questions? Did you look up any positions? But that's what I have them do. Or are we doing it together? We find some positions and I put together some questions I think they'll get. And then we try to put together some some succinct answers, right? Like to questions. You know, so they can memorize them and don't memorize like a script because people can tell. And I'm sorry to all the people who are interviewing English and English is not their first language. I'd rather hear you be spontaneous and not perfect English than like a robot reading something because when you're interviewing someone, you want to sense their emotions. So I'd rather have somebody in choppy English explaining something that happened on their data science project than somebody just saying we did this project that was a dashboard at a hospital. You know, whenever anybody does that, I mean, they even seem kind of suspicious. It's like, what really happened? I always want to ask what really happened? You know what I mean? And so it's better if you're saying, well, you know, I tried this, we tried that, we were our goal was to display this, you know, and just have it be a spontaneous answer each time. But you might have talking points like, again, thinking with my intern, I made some dashboards with her. One of the dashboards we made, I pretty much designed what I thought we should have on it, but she built the whole thing. And I didn't totally design it, like I told her what I wanted, and she told me what the limitations were, and we sort of worked it out. So if she was interviewing about that, she want to work on her talking points about what she did, and she did a lot. Like, I don't know how to program a front end, and she did it. Actually, let me put it in the chat so you can see it. I'm so proud of her. Just because it was just a demonstration project, like she's, because she's listening to me. She's so awesome. And she's doing a few of these portfolio projects. Here's an example of her. I mean, it's technically ours, dashboard. And also, you know what, let me give you the link to our peer reviewed article on it. I should, of course, prepare these things, but I never know what I'm going to bring up because I just talk and talk and talk. Let's see here. Where did I go? Here it is. So let me go to the article. Of course, it's slow. So this is a peer reviewed article that I'm giving you a link to now. And actually, let me just change what I'm sharing here. And just do a quick shout out to why peer reviewed articles are important. So if you guess that my intern is one of my favorite students or whatever, you're right. I just love her. Her name is Natasha. You can see she's the second author of this. Okay. So why do we have this peer reviewed article? This is what we do in academia for quote unquote portfolio projects, right? Like the best thing you can do into academia, if you do data science, is to publish in the peer reviewed literature. Now, if you want to, hi, nice to meet you in the chat. Thank you. Now, if you want to publish in the peer reviewed literature, it's actually not that easy, right? Like, if you've ever heard the term publish or perish, that's said by people in academia, because if you become a professor, you have to do this, you have to publish in the peer reviewed literature a lot. And actually have another blog post on academic publishing and trying to get into academic data science. It's a slightly different path than a path where you get into like solution space data science, where you're like making software, implementing AI. And my most of my life has been spent in academic like data science. But even, you know, I'm a curious gal. So I've been involved in software development. You know, I've done a lot of stuff, obviously, I've made a data warehouse before. But like, for example, I made a data warehouse, I ran a data warehouse. But I was also publishing in the informatics literature. I was already I was doing studies on my data warehouse. A lot of people don't do that. And then they suffer in as data scientists, their career suffers. So you always need to be doing these portfolio projects, which are peer reviewed papers. And if you choose an academic pathway in data science, I say you have to, you don't have to, but you're at an extreme disadvantage if you don't. And you might notice that Natasha doesn't like, well, I don't know if we couldn't we didn't have to say our degrees. But Natasha has she has degrees in music, you know, that's her previous profession, which she still does. But right now she's working on her master's in public health. I think that's it. I can't remember. Spiostatistic, something like that. And that's going to be her data science really degree, right. And so it's even before she's finished with that degree that we published this. And that's why because I love her. I love her so much. I love her so much. And I want her to succeed. So I'm doing this to her. Oh, Puneet, thank you. This is so nice. Puneet says you bring a lot of death. I'm going to talk. Thank you. Let's see if I can get better at the software. So I'm talking a lot now about academic data science, partly because I know about it and partly because I'll just confess to you like actually I meant, I'm trying to get more customers. So I met a data scientist. Well, actually, he wasn't a data scientist. He was just a plain scientist. Okay, he was from like physics or something. And he wanted to go into data science. And I told him he's like, if I were to hire you as a paid mentor, what would you do for me? And I literally looked up in his country, they had a health surveillance data set and he wanted to go into health data science. And those of you who take my LinkedIn learning courses know I demonstrate a lot using the BRFSS, which is the behavioral risk factor surveillance system health survey data set that we do in the US for health, right? And so I he's not in the US. And so I wanted him to get a job in his country, right? So I found that his country had a data set similar to that. Of course, I didn't know anything about it. But I said, if we were going to work together, what I would recommend he do is we do an analysis of that data set and publish in the peer reviewed literature. That way he could go from like physics where I guess he had published in the peer reviewed literature. But it didn't really count because it wasn't in health science. And so I'm like, well, if you want to shift right away, like you don't want to go to zero and start over, you just want to shift right away. I would just do this project with you. And I think you would learn everything because he already had done us data science certificate, he'd already taken courses and stuff. It's just that being able to apply all that. That's what actually everybody's telling me is being able to apply what they learn is the hard part. Like even my intern is brilliant. This application is what's hard. She can learn anything. It's just the application. So long story short, if you ever hire me to try to help you with this portfolio project stuff, if you're in more of a business situation, I'll try to do something like I did with the Massachusetts casinos. And if you're more going for health data science or academics, I'll do I'll give you the intern treatment. All right, so this is what you'll get. So let me see here. So let me go back to this. I'm learning how to use this. Let me go back to our blog post. Of course, I forgot what I was talking about. And yeah, so tips for effectively answering interview questions is that's where we are. So you can imagine that if my intern ever wants me to practice with her, I actually don't think she needs practice because she's advanced, you know, like she already had a career before this, you know, people who've had a bunch of careers or a bunch of jobs, they're better at interviewing because you just have to do stuff in your job. You're so good at talking about it later in life. But if you're new to interviewing, or you're just shy, you just don't have a personality that's good for that, you really want to practice delivering answers. And actually, the third point I make on there is you should be just as comfortable answering a positive question as a hostile one. Let me actually tell you, since we're having this conversation about an interview I had at Blue Cross Blue Shield. And I'm telling you who they are, because this was so long ago, this was in 2011. So anything I tell you that happened, I'm sure it doesn't apply anymore. But it was in Boston. Now, I just want to give you the setting is at the time I had this interview, there was a lot of innovation going on between insurance and the state. And literally Blue Cross Blue Shield, their building was on one end of Boylston. And the state house was on the other hand of Boylston Street. And they would go down to the state house, that's our capital, and find out what legislation was happening, and then come back and try to execute it in the afternoon. Like I felt really sorry for them. So the fact that I'm going to tell you about probably the most dysfunctional interview you'll ever hear, I'm warming you up for the fact that our healthcare system was super complicated locally at that time with Blue Cross Blue Shield. And actually, if you go to my YouTube channel, you'll see one of my most popular videos is about healthcare reimbursement. And that'll kind of help you understand why me applying at this insurance, they were kind of acting weird. So I went to, it was a headhunter, right, like a talent company. If you look at my blog post about the different kinds of companies that interview. So this headhunter, I connected up with her, and she had been hired by Blue Cross Blue Shield to hire data analysts, because they had a lot of data analysts in different positions. And it was hard for them to keep those positions filled. So they had this headhunter going around and finding people and interviewing them for positions. So on one hand, that was good, because I really got to know the headhunter. And she knew everybody at Blue Cross Blue Shield, and she was really, really good at preparing me. And she's like, Monica, this is going to be an eight hour interview, it's a day long interview, like you go there in the morning and you come home at night. Luckily, I lived like a few buildings down. Like I kind of didn't even really want the job, it sounded bad. But I lived a few buildings away from the office building. So I was like, I'll try it. I don't have to go anywhere for the interview just really close. So I said no problem. I'm have no problem getting the job interview. I live right by the building. And so she was explaining how the reason would be an eight hour interview is how it was structured is there was going to be another data analyst who was sort of new there, who was going to take me around to escort me to interview with different big wigs, I guess, leaders. And they were each going to ask their questions. And then we're going to go to lunch. And then I was going to meet with who would be my boss if I got hired or who would be the leader. And that would be in the afternoon. So I was like, all right, you know, and she prepped me and all that. She's very nervous about me. I don't know why, but anyway, so I got to the interview. And the first thing I learned is that the data data analysts, the newly hired data analysts had missed the train. So he didn't get to work on time. So there was no one to escort me around. And no big deal. But for some reason, that completely derailed the interview. Like, the fact that that happened, suddenly nobody knew what to do. So that was a red flag. Okay. So I don't remember much. I'll tell you the interviews I remember. And one of them is this third point on this blog is, and this wasn't the first interview I had that day, I talked to different people. But I remember I walked into this room is like a little conference room. And there was this woman seated. And I didn't know who she was. Like, I normally people introduce themselves. They start talking. But she just yelled at me. And she said, she was like holding my resume in her hand. She's like, why would I hire you? You're an academic. I'll just, you'll just leave after a month. You won't stay. You know, like she said, you know, so, um, as I say on the block, what my talking points is that I'm compassionate and kind. And I just actually felt sorry for her because I was a manager. And I know if you're a manager and you're doing that at the interview that you've kind of been pushed past your stress level, where you're acting inappropriate. So I said to her, you go, you know, maybe you're right. Maybe I'll leave this job in a month if I take it. But let's just talk about it. Because, you know, if I just figure out I'm not a good fit for the job, then I just won't take the job. And then you don't have to worry about it. And she kind of calmed down, you know, and but okay, so that was her. Then the guy showed up who had missed the train and we went to lunch and he looked depressed. So that was another red flag. And then there was oh, there was this guy who his office had a desk on it that was just piled with papers. And he had this organizational chart out and he couldn't figure out what what position I was interviewing for on it because there's so many data in the analyst positions. And I noticed when I looked at the chart that there were no women at the top, there were no women leaders. So I mentioned that to him and he's like, Oh, there's one she's on maternity leave, which is the wrong answer. That's the wrong answer. So in fact, that guy I think was filling in for her while she was on maternity leave, because I said to him, I was like, he, you know, I said, I like that you feel so comfortable that you have this messy desk in front of me, you know, I always have a messy desk. I'll just be honest you I'm a messy desk kind of person. And he's like, Oh, this isn't my desk. This is a temporary desk. I have a real office at some other location. So it was like total chaos at this place. And then they offered me a job like I say on the blog post, but I didn't take it obviously would have been an awful job. Probably that lady was correct that manager that I would have left it in a month, but I knew myself better than that, which is why I say there's a live stream we just had where I was telling people, you got to research these companies before you start working there. Now, I knew a lot about Blue Class Blue Shield, but they're a huge company. So who knows, right? But it's still if I was against something that they did in general, I wouldn't want to even look at them. Okay, so those were the three main tips, which were to have a bunch of portfolio projects so you can say something you know what to talk about, and then practice putting together really succinct answers, and then practice interviewing with somebody and have them try to trip you up and stuff. And also, one thing my customers often will worry about is they'll try to say everything about a project they do. And I tell them, no, it's better to just kind of answer the question. Like, usually the last question is like, here, I'll stop sharing my screen, like, tell us a time when you disagreed with your boss and how did you handle it, right? And oftentimes, what I would do is I just run some data, like if the boss said this, you know, like, with me, I, once my boss said that SAS and SQL ran equally as fast. So I set up an experiment and we found out what their thresholds were, you know? So notice how I answered that question in just a few sentences, even though we did a ton of work on that threshold study. The reason why it's kind of good to do that is because then the interviewer can ask follow-up questions. So I don't know what the interviewer is going to be interested in. Maybe they're going to be interested in the study design. Maybe they'll be interested in the fight I was having with my boss, you know, like, why were we arguing about that? You know, I don't know what they're going to ask about. So I don't know what they're wondering about. So it's better if you give sort of, like, don't make the answers too short. You know, make sure that they have enough information in them so that they are, like, a little story, like, a little vignette. And then they can ask. Like, sometimes I tell a little story and explain. I give you some details. But then you can ask more questions. Like, you could ask me more about what happened at that Blue Cross. Like, I call it my Saturday Night Live interview. I felt like I was in the SNL skin. Those of you who watched that comedy show. But, yeah, so one of the things that people sometimes have trouble with is coming up. What are the interview questions? And you can Google for some. But after a while, if you actually go to interviews, you'll kind of know what they're asking. Oh, good. I'm glad that you're liking this. I like happy faces. You guys have got to, you guys have got to ask me some questions because I need to know, you know, what exact topics you're interested in. You know, I have another blog post I should probably show it. Let's see here. I'll look it up. It's about how to research data science companies. Well, not really data science companies. Just companies you're thinking about working at. I put a few of these blog posts together with these live streams because I realize these are what everybody's asking me about. So I might as well, you know, just come up with like some better answers. Okay, so let me share the screen with you. And I think Joe, you can speak up if you agree, is I think Joe thought this was really important too, which is looking at companies you're applying to. And so, and I did a live stream on that I'm not, you know, I'm getting better at publicizing these, I recorded it, you can go see the recording if you want. And so, first, I mentioned that there are these recruitment talent companies like the headhunter that worked with me. So what happened was she came to me and I said, I'm not even talking to you until you tell me what business you're talking about, because I don't want to waste my time. Well, I was a manager before and she was trying to hire a kind of a higher level. So she pretty much had to spill the beans, right? But sometimes what will happen is these headhunters are hiring at lower levels, and they'll have a lot of customers like a lot of different businesses, and they'll be kind of like dealing you around shopping you around a lot of places, and they don't want you to know all their places. So sometimes they'll be pretty quiet about it. But your goal is to find out who they're shopping you to specifically, because you want to be able to research that company. And then, let's see here, you have to turn my mic on. Oh, Joe, I have to have you as a guest to interview if you want to talk. We should set that up. So everybody, I'm going to set up an interview with Joe. I'll connect with you. What happens when I interview you is like I do like this and you're the thing over there, right? So let's do that one of these days. And then I'll publicize it and then we'll be able to hear from you. I'm so sorry. Let's see here. So, oh, Farhad has a question. Some of my friends who are working on data science are telling me, if I'm going to work on data science in the future, I do not need to study or work on research articles. I just need to do coding and so on. Okay. Well, this is the answer. If you go into academia, if you go into academia, like you go work at a research institute or you work at a college or maybe some places in the government, you will have to write reports. They will either be peer reviewed reports like that one I just showed you with my intern, or they'll just be reports that are kind of like white papers. But you're going to have to do that as a data scientist, just like Natasha did. She has to be on it. Like there's a group of people. And if you're not in there and you're just in business, you are at a distinct disadvantage if you cannot make a research do a portfolio project on your own. People who just code are programmers. A data scientist is supposed to have a little bit more just like more ability. Like a lot of people can learn programming languages. But if you take my study design courses on LinkedIn Learning, my data curation course, you'll say, oh, that's not programming. And that's the kind of stuff I need to know for data science to put it all together. It's like this other glue. You need to glue your skills together. So I don't know what's up with your friends, right? Like your friends, you could go become a SQL programmer somewhere and say you're a data scientist, be perfectly happy. And if that's what they're doing, then maybe yeah, all you need to know a SQL. But I'm not sure. When I think of data science and hiring a data scientist, I'm not thinking of hiring somebody that I'm going to tell them what to code. I'm thinking of hiring somebody that's going to be able to figure out something like that Massachusetts casino portfolio project example I did, or something like that, or like that dashboard that my internet I did. And so let's see here. Oh, thank you. I'm glad that you have a positive response here. So that was, now we have a junior he asked, when researching companies, what's more important? Company structure mission, team functions or role responsibilities? Well, actually, what I tend to do, like, Joe asked, what's important to you? And I like that he asked that, right? Because you're actually I'm I made a video, I haven't published it yet, but check back. And in the video, I'm showing you I'll show you how I think of these companies. And actually, if we go to the website here, you'll see how I kind of did it backwards, Junhee. I started with an announcement for a position like let's say I wanted to apply for this. So if I have learners or people applying for jobs, I tell them, bring me a few position descriptions that are like what you want. And let's say one of them they really like, but we're not familiar with the company. This was we're not familiar with company. I looked up the company and I showed on my blog page. And in the position description, it describes some stuff about the company. And you know, somebody wrote that position description, somebody wrote that communication about the company. They wanted me to know that, you know, so I started looking for like the founders profile. There wasn't even a web page for this company, but I found public records about it and news about the company. So now I started getting a feel for the company. And when you see the video I post later, you'll see I do a little of that in real time where I go to two different web pages. And I just look at their menus and I talk about how that communicates their values. Now, what Joe said here, which I think is really important is data scientist equals forensic scientist. And I'll I'm going to pitch my data curation course, because curation helps you do that forensics. What you're trying like, it's hard. It's kind of like, imagine you were hiring somebody at a detective agency. And you ask them in their interview, our detective agency filed for bankruptcy one year. What was that? And they were like, didn't know, because they didn't do any detection. You'd be like, I'm not going to hire a detective like this. And so data scientist, it kind of goes like that, it's kind of a joke, but you'll be like, you know, what kind of investigations did you do? Like, I don't know, you know, like, I guess your company's okay. You know what I mean? Like you kind of would hope if you're hiring a data scientist that they come in, you know, going, Hey, you know, I saw you just released this drug Pfizer or whatever it is, you know what I mean? Let's see research. Farad asks, thanks a lot. Research articles and CV important to data science companies, only two companies in academia. So let's say that there's a company that develops drugs, like, like, Genentech, I'm just guessing, let's say you applied at Genentech, if you had those research articles in peer review publications in your CV, it would look good. But let's say you applied it like Siemens or like a cybersecurity firm. They would look more for like I triple E like engineering white papers. Now, what's the difference between a peer reviewed article and a white paper? White papers are easier to write because they don't you can just write one and post it like I did on my casino one. And they there's no formal thing or whatever. And they're very technical. So they're not very formal. Okay, nobody's really doing oversight. But they're technical and you can just say what you want, you can just explain everything. Now, it's interesting because if you go to that dashboard that my internet I made, and then you read our peer reviewed article, it's very heady, it's very philosophical. Okay, it's not a white paper. So they have different purposes. And it's harder to publish in the peer reviewed literature, because it's peer reviewed, people are reviewing or whatever. So it's way easier to write a white paper. Now, if you're in health or academia, and you write a white paper, there's nothing wrong with you. But they're just they just look oh, they're cute little white people. Look at the engineer wrote a white paper for the other engineers to read so they can play engineering, you know, that's sort of how they see it. I mean, personally, I don't, I read SAS white papers constantly, constantly. Okay, so I don't see it that way. But a lot of academics kind of grow into that attitude because it's so hard to publish in the peer reviewed literature. There's just different audiences. And then, Joe adds CVs are very important. It helps you prepare for interviews. And also, yeah, and I would add it helps the person interviewing you prepare for the interview. So because they're going to ask questions and see how good you are at answering and to get back to what this is just a good statement here, data sciences forensic scientists, when I do forensic interviewing of people, like I think they committed a crime, and I'm interviewing them, I already have all the evidence. I just ask them about it and see if they know it. So it's kind of like, if you're, you know, on TikTok with a bunch of dances, and they ask you, Oh, do you like to dance? You know, if you say, Yeah, I'm all over TikTok, you know, they, they, they already know that, right? Because I should already know that if I'm a data science manager, I should have figured out you're already all over TikTok, you know, because it's just data, it's public data. And so, so yeah, so if I'm, like, if I read your CV, and it says you worked on such and such study, or you worked on such and such data warehouse, or you work for such and such project or initiative, you know, because sometimes you have these big initiatives that are multi groups worked on, you know, I'm going to maybe Google it and look up that initiative better understand it. Like if you're a data scientist and you're applying in health, and you used to work in like insurance, which is related, you know, I might need to like, learn about insurance to interview you. So yeah, your CV is really important. And oh, that's another thing I do. Maybe I should post about it. I think I have a blog post about it. I'm super picky about CVs. Like if I have customers, they're like, look at my CV, I'm like, throw it away. I'm starting over. They start crying. I'm so rude, because I think there's only one like format for CV. And just to be clear, this is what a CV is. And this is what a resume is a CV stands for curriculum vitae. It's a document and it's like a resume that's used in academia. But what's different about it is it's long. It lists all the academic things you've ever done. There's not an official list. Obviously, you have your positions and your education. But if you're like a clinician, you've got your licensing. If you're like me, you're publishing all the time, you've got all your publications. If you do a lot of presentation at conferences, that's on there, all your academic things. And as you get older and do more stuff, it gets longer and longer. So the CV gets big. And if you're applying for an academic job, you turn in your CV. If you're applying for a non-academic job, like when I apply for non-academic jobs, as a data scientist, I have a resume. So what's a resume? One or two pages. And it's really focused on summarizing the projects you've done. In fact, I talked about this in the last live stream about how I have a friend who has an IT business. And I don't do a lot of contracts for her, but once in a while, she gets a data science contract and I do a contract for her. If I ever do that, sometimes her company wants my resume. So I just throw one together and I talk about the little projects I do. Each time I do one for her, I put it on there. Joe says, the federal government jobs require a CV that provides details of your work history. That's true. I don't know as much about their format. I know NIH has an NIH-formatted biosketch that they tend to use in academia. That's like, if you're an academic and you're like, I don't want to give you this 20-page CV, you tend to use an NIH-formatted biosketch. I could say that 50 times fast. NIH-formatted biosketch, like we say, and that's kind of what we use. And so what Joe's talking about, the federal government, a CV that provides details of your work history, often they'll be templates from the government. So if you're in data science, you're like, I don't need a CV. I'm going to do a resume. Well, you can often find templates from governments about how to put your resume together if you're going to do a contract for that government, just to let you know in case you do government contracts. Like, I always end up. Well, we're two minutes before the stream ends. And I wanted to thank everybody who showed up and asked questions. And I want to thank you for showing up, Joe, and sharing your wisdom with all of us. And also just a question, I'm getting, they're good questions because I'm not actually looking for a job. I'm helping people look for jobs, and I'm helping people improve their careers in data science. And so I need to hear from people, the kind of people I'm helping. And so I really appreciate that. This last stream, it's going to be available on LinkedIn at the same event link. But I also have this YouTube channel that I'm trying to grow. So it'll be on YouTube. And if you go there right away after this, it won't be anything special. But I like to put more in the description. Every time I give you a website, I like to put that in the description so people can go back or you didn't have to take all these notes. So please, if you enjoyed this live stream, and if you thought you got something out of it, please go and subscribe to my YouTube channel. And you won't be disappointed. You'll get some good data science resources and feedback. All right. Well, thank you very much for spending some time with me on the Saturday. And I really appreciate everybody who showed up. And if you're looking for a job in data science, good luck on your job search.