Data science roadmap: What skills you should learn first?

Hi everyone welcome back to another youtube  video in today's video we're going to talk about   what are the first things you need to learn  if you want to become a data scientist   data scientist has been a term that has been  thrown around in the industry specifically in   the tech field in the recent years especially  after the harvard uh harvard study called it   the sexiest job in the 21st century uh which is  debatable and definitely a topic for another video   but you can see that it has gotten so much  attention because of that and also there are   obviously like some good signs to be good sides to  being a data scientist so because of that there's   a lot of information on the internet some is good  some is okay and some is just does not make sense   at all so i am going to share some of the things  that have personally helped me transition and what   i have learned from my experience for those of you  who don't know i come from a non-tech background   and i became a self-doubt data scientist i went  to business school then transitioned to becoming   working as a data engineer and then became a  self.data scientist um if you have not watched   in my previous video where i talk about my  journey i'm linking it here so make sure to   give this video a watch and see you might find  it useful so before we jump into the video make   sure to give this video a thumbs up to help with  the youtube algorithm and subscribe to my channel   if you would like to hear more from me on the  topics of data science as well as career related   uh topics so talking about uh what is the first  thing you need to learn well let's step back a   bit so in terms of becoming a data scientist first  thing i would suggest to figure out data science   is an umbrella term and the term data scientist is  used very liberally that it means it does not mean   the same thing to one person versus what it means  to the other person and the same stands for the   companies that are hiding for data scientists in  some companies data scientists by data scientists   they mean applied scientists a person who is a  developer times data scientist in some companies   a business analyst is considered a data scientist  so um keeping that in mind what i would suggest   you to do is first figure out where exactly i'm  going to link this video here where i show that   diagram in terms of what exactly you need to do  look at that diagram figure out where exactly do   you want to be on that diagram do you want to be  in the middle which is kind of more of like like   a generalist data scientist then there is more  technical which goes into applied scientists then   there's a machine learning side and then there's  more at the bottom there's more data analyst focus   or business analyst focus so first thing you need  to do is figure out where exactly do you want   to be on that venn diagram and once you figure  that out i promise you your your your path your   your journey is going to be much simpler because  if you don't know where you want to be on that   diagram you're going to be all over the place  you're going to be learning statistic you're   going to be learning machine learning you're going  to be learning python you're going to be learning   tons of other things believe me the data science  field is huge and you can easily get lost   learning trying to learn everything so if you're  able to narrow down your focus able to narrow down   what you want to be and then work backward it's  it's only going to help you so let's say you have   figured that out um and you want to uh you want to  be a generalist you want to be a journalist data   scientist which is basically somebody who has good  business knowledge has a good statistics knowledge   good machine learning knowledge and  understanding of software development   process and is able to combine as well as  product management and is able to combine   all these together and apply their knowledge to  solve business problems so you have three options   one is you can go back to school you can enroll  yourself in a data science degree program   and you can follow through that program that  curriculum and then graduate with a degree and   build some projects and then look for a job the  other option is the second option is you can   do that through a bootcamp bootcamp is a version  it is a it is a i'd like to think about it   in a way that is the degree program but  um it teaches you everything that a degree   program teaches you but in a much condensed  fashion um and that's why those programs are   some normally like eight to twelve weeks and those  are intense time periods because you uh intense   programs because you have to learn a lot during  that short time period so boot camp is another   is another option that it requires a lot of your  attention so for a degree program you might be   able to do it if you're already working um so you  might be able to do that part-time finish that   degree program for a boot camp i feel like you  need it needs your it needs your like dedicated   attention to be able to get the full advantage of  the program so um those are the two options the   third option is which is something that i did and  a lot of people do end up doing is going the self   teaching route and that is through learning um  through like bunch of courses learning on your own   uh from peer mentoring from doing the work  and all of that everything combined watching   youtube videos um and i am primarily going to  be focusing on that last bucket today in terms   of the self learning and for those of you who are  trying to do that what are what is the first thing   you actually need to get good at to be able to get  get your journey started um so for me personally   what was super super helpful and i would highly  encourage you to consider that is uh starting with   and i know a lot of people start with python like  there's an advice out there that says let's start   with python some advice says that you go and do  r some say is like go learn sql i think those   are absolutely okay but those are that's okay  advice but let's step back a bit sql python r   is a tool to apply data science it's not the data  science by itself so if you really want to learn   the knowledge learn the field i would suggest you  to start with statistics for those of you who are   already working as data scientists you would agree  to this that the statistics is the fundamentals of   any data science rules whether we're talking  about a data scientist generalist or we're   talking about a machine learning scientist  or we're talking about applied scientists   so statistics are the fundamentals of any data  science role so the first thing i would suggest   you to do is start with statistics and once you  have figured that out then you can add more to it   whether that is python or whether that is going  more deeper into machine learning a skill set but   like starts with the basics start with statistics  and when it comes to learning um learning into how   do you learn that there are multiple ways to learn  if you have taken statistics as part of your uh   schoolwork like i think that's pretty much enough  there is not much to statistics it's just the   basic fundamentals and it doesn't change much so  whether you learned it from your curriculum from   your course that you have taken at school or you  learn from a book or you watch a youtube video and   like combine all those concepts together i think  anywhere you can learn statistics can be learned   anywhere because the concepts do not change the  concepts the statistics concepts do not change   the teaching style changes from where you're  learning from so i get i guess like it all comes   down to like your learning style how you learn  best whether that is through video content whether   that is to reading or whether that is through  taking a course so figure out what what works   for you and then start with statistics build that  foundation and once you have built that foundation   then you can go to the next step um and we can  talk about it in the next video but then you can   start um learning the coding languages and  there's a i have a lot more to say on the   coding languages uh which one to get started at  and it's a slightly controversial topic because   different people have different opinions on what  language to learn um i definitely have my one of   my own so i will definitely share that in another  video but um in terms of learning statistics   yes definitely start there for me what was  helpful is um i started learning statistics   one of the books that my manager recommended me  at that time when i was working as a data engineer   is let me show you so this was and i promised this  is not sponsored sponsored i just found it in my   cupboard um and i was like this is perfect for  this video because um this is something that i   did use when i was learning uh statistics so this  is a book practical statistics for data scientists   uh by peter bruce and andrew bros so you  are welcome to buy it from anywhere you want   and learn statistics that way um and then in  addition to that you can watch youtube videos   there's a lot of ton of youtube videos um on  statistics that and khan academy is a great   resource for those of you who haven't taken  i haven't i'd be surprised if you haven't   pursue ever if you haven't taken a video on  khan academy but it's a great resource for   uh and it does a great job simplifying  complicated topics so what i normally did   is i got this book i did not read everything  um i would start like initially with a concept   then i would read about it in this book i would  read a blog about it and then i'll watch youtube   videos or khan academy videos which simplify the  topic for me further because i feel like learning   it just like reading through it one time it's a  little bit harder i just i just want to see like   how different people would explain it differently  so that's what i would normally do is like like i   would read it here and then i would go watch a  youtube video and i learned a topic for another   video again i know i keep saying that financial  literacy i became financial literate last year   and that exactly was my route i would read it  in the book then if i needed to build further on   that topic i would go to youtube or i would read a  blog and things like that to further my knowledge   so hopefully this was useful i know there's a  lot of information out there in terms of how   to get started this is personally in this is my  opinion that in in my opinion and in my experience   you need to start with statistics um and then  you can add on top of it because statistics is   the foundation of data science so hopefully this  was helpful for you if you have any questions feel   free to leave it in the comment or if you have  any feedback for me or ideas for future videos   feel free to let me know alright i'll talk  to you in the next video have a good one bye

Channel: Sundas Khalid

Views: 338,365

Keywords: data science, data scientist, self-taugh data scientist, tech jobs, amazon, women in tech, big tech, google, statistics, machine learning, sql, data analyst

Id: ylOILe-Sc-w

Length: 10min 36sec (636 seconds)

Published: Fri Aug 06 2021