 Now on the surface, it may seem that being a data scientist is all sunshine and rainbows, at least that's the vibe I hope you get from my videos. High pay, great benefits, flexible hours and interesting work are just some of the key things that come to mind when you think of a job in data science. While these are definitely true, every job has some struggles behind the scenes and data science is no exception. Don't get me wrong, it's a fantastic job and I absolutely love the field but not every day is completely glorious and euphoric. That's why in this video, I want to go through several of the harsh realities that I face as a data scientist and some that you may face as well. This is not to put you off becoming a data scientist but just some of the things or feelings that you may expect to have over the course of your career and hopefully it will help you decide if data science really is a choice for you. Let's get into it. Unless you've been hiding under a rock, it's no secret that last year AI took the world by storm. Until 2023, most people would simply roll their eyes if you mentioned the word AI but now Chatchapiti is a household name. The rise of generative AI is just one example of how dynamic the field of data science and machine learning truly is. Back when I started as a data scientist in 2021, there was no texted image generation using Dali or github codepilot to help you write cleaner code or even Chatchapiti as a general AI productivity tool but now these are my go-to tools that I use on a daily basis. The problem is as a data scientist, you're directly involved in this field so you kind of feel obliged to keep up all these latest trends and developments so you're not behind the curve. This can be tough especially if you haven't got much time outside of work hours to upskill in these key development areas that are booming at the moment. Obviously the popularity of generative AI is an extreme example and it's kind of like a revolutionary technology. However I can give you one a bit closer to home. Last year I carried out a project where I had to migrate one of our models to a new package because the existing one didn't work on M1 Max. I won't go into all the details but basically the existing package didn't install on M1 Max because the chip architecture is different to the regular Intel ones you get at most computers. Anyway halfway through the migration the existing package or the old package that we used for our model released M1 Max support so basically all the work I did was made completely redundant. As annoying as that was it does going to show you that data science and tech in general is such a rapidly moving field and it's hard to stay on top of everything all the time. You literally can't complete data science there's so much to learn and also there's so much research going into it that it's got to keep on evolving so there's always going to be something new for you to understand and delve into if that's what you want to do. Now that can be both scary and also very fun because like I said you can't learn everything and so if you want to be the type of person where over time you slowly accumulate all this knowledge and there's nothing left for you to learn then data science is not the field for you but if you're more like me and I love continual learning then it's probably the best field for you because like I said there's always something new to delve into or a new area or a new skill that you really want to pick up and develop that it'll take several lifetimes to really learn everything and even that other thing is enough. As a data scientist you have to accept that you won't learn everything you may be aware of everything but it'd be impossible for you to know every single bit of the field. With this constant change in the field imposter syndrome is very real for data scientist. I feel an imposter at least once every two weeks and while I can't speak for other practitioners from my anecdotal conversations others often have this feeling from time to time. By immediately opening up LinkedIn or Medium you're bombarded by people building these cool projects and these fancy models to solve weird problems. Someone may be doing you know a deep reinforcement learning agent to predict the weather or a chat GPT time series forecasting algorithm to measure stock price returns. Now obviously these are just fictional examples but they are based on some ground truth that I see frequently on these platforms and it's quite difficult not to look at all these cool things people are building and not think that you're behind. It often feels like people are creating things all the time whilst you're not keeping up with them all your knowledge is not as in depth as them in that certain area. It's important to realize that this is not the case and that you know people aren't constantly creating and building things all the time while some people are like this the majority of data scientists are not. One data scientist may be really good at reinforcement learning but have really weak knowledge on combinatorial optimization or vice versa. What you often find is that people really only speak about things they're really good at or they know a lot about and less about their areas than they know least about. I mean that's just common kind of human nature you want to speak about things you know about nor the things you don't know about and that's why online it may seem like people are doing so many things but they're not really they're just doing one thing and they may release one project you know every two months but because the internet is filled with billions of people it seems like people are doing things all the time but they're not it's just a platform of showing you those things. The main point is that if you want to be a data scientist be prepared to feel like you don't know enough or that even like your failure sometimes but remember that's not the case and I promise you a lot of data scientists out there definitely feel like this from time to time as well so you're not the only one. Now data science is quite a new field and the term is still not very clearly defined and this often leads to inconsistencies with job postings the people you speak to or even if you work at company people may not know exactly what you do as a data scientist. If you are a divorce lawyer it's very clear what you do likewise if you're a doctor or a dentist however the data scientist is way more diverse and people have so many opinions or at least views of what you actually do. A data scientist at one company may be completely different to a data scientist at another company at one company you may be a data scientist that does a lot of machine learning however at another company you may be a data scientist that does just completely analytical work again it varies so much it also depends on the structure of your company if you're the only data scientist at your company then you can expect to be doing things from data engineering all the way to the analytical side and even maybe a bit of software engineering depending on how many developers you have at the company this ambiguity around data science often leads people to not understanding what you do and not utilizing your skills correctly this can be both good and bad the good is obviously that you learn a lot of useful skills away from software engineering to analytics and the business side the bad is that it's quite hard to find a job that aligns your skills because like i said at different companies you may develop certain specialties in certain areas but this data science specialty you know one company may not translate to another company so it may be quite hard to find a job in that respect however over time as data science becomes more established i feel like this become less of an issue as i'm sure you know the fundamental skills for data science are statistics maths and programming now on their own these skills are quite tough to learn but as a data scientist it's expected to know all three of them to a good standard in other jobs extending at least one of these skills is one of your key standout strengths but in data science this is far from the truth it's the baseline that most data scientists have so you have to have some sort of extra dynamics or an extra thing that you're really good at to really stand out from the crowd this sets the bar very high for data scientists and so there's a really high entry requirements just to break into the field and also be really good in your profession like i said you gotta have some other skills that makes you really stand out everyone has good maths statistics and programming skills so what can you do that really makes you different from the other data scientists out there there are a few things you can do to stand out such as be a specialist in one set of domain so you're the go-to person when any problem in that area comes up people will come to you to solve that problem you can even go the other way and become more business focused so you really know your domain well and how your company functions so you can quickly solve the business problems and have in-depth understanding of how the whole operation runs and another way is that you develop your soft skills so you're a great communicator writer and you can explain technical concepts in a really digestible manner and that's again a very useful skill set to have particularly when dealing with non-technical people now the problem comes in in that these extra skills outside your normal job take time to develop and it can be quite hard particularly if you haven't got time outside of work hours to work on them if you want to learn the required skills for data science machine learning i have a previous video which i'll link on screen here that'll tell you the roadmap i would do if i was learning machine learning from scratch again now i hope you enjoyed this video and i hope it shed some light on some of the problems i face as a data scientist and i really hope it made you decide if this career it really is for you because these are some of the key principles or you know realities you have to accept if you want to be a data scientist if you enjoyed this video and i see more content like this on this channel then make sure you click the like and subscribe button and i'll see you in the next one