- 34 minutes ago
2024 is shaping up to be a major year for data, and this CareerFoundry webinar breaks down the five emerging trends that are already changing how teams work. Alex The Analyst shares his perspective on AI, synthetic data, hyper-personalization, freelancing and consulting, the cloud divide, documentation, governance, and data unification, with real examples from startups, enterprise teams, healthcare, and finance.
The discussion goes beyond buzzwords and looks at what these shifts mean in practice: AI built into everyday tools, companies using synthetic generated data to test scenarios, more demand for independent data experts, and a growing push toward better data security and compliance. It also explores why some businesses are reconsidering cloud costs and moving toward on-prem servers, especially when downtime and long-term expenses matter.
If you are learning data analytics, exploring a career change, or tracking the latest data trends in 2024, this webinar offers a clear, practical overview of where the industry is heading. It is especially useful for anyone interested in AI in data analytics, cloud computing, data governance, startup consulting, and the future of data teams.
SEO: Ideal for viewers searching for data trends 2024, emerging trends in data, AI in analytics, synthetic data, cloud vs on-prem, data governance, documentation, freelancing in data, and data analytics career insights. Great for webinar replays, career development, and industry analysis.
The discussion goes beyond buzzwords and looks at what these shifts mean in practice: AI built into everyday tools, companies using synthetic generated data to test scenarios, more demand for independent data experts, and a growing push toward better data security and compliance. It also explores why some businesses are reconsidering cloud costs and moving toward on-prem servers, especially when downtime and long-term expenses matter.
If you are learning data analytics, exploring a career change, or tracking the latest data trends in 2024, this webinar offers a clear, practical overview of where the industry is heading. It is especially useful for anyone interested in AI in data analytics, cloud computing, data governance, startup consulting, and the future of data teams.
SEO: Ideal for viewers searching for data trends 2024, emerging trends in data, AI in analytics, synthetic data, cloud vs on-prem, data governance, documentation, freelancing in data, and data analytics career insights. Great for webinar replays, career development, and industry analysis.
Category
📚
LearningTranscript
00:00:02Hi, everybody. Welcome to another Career Foundry event this evening. I know that a lot of people
00:00:07are joining. Drop us some emojis because we've got an emoji bar at the bottom. Also, we've got
00:00:13a chat on the right-hand side. So whilst people are joining, just drop your name and why you're
00:00:19interested in data analytics. We want this to be as interactive as possible. Alex wants it to be
00:00:24as interactive as possible. Obviously, the emoji is coming in already. Great, great job.
00:00:28So don't be shy. Yeah, just drop your name in the public chat. And if you're watching on YouTube
00:00:32or LinkedIn, because we're streaming there too, also drop your name there too. This evening,
00:00:37it's all about the top five emerging trends in data in 2024. We are recording this session. So
00:00:43if anyone misses it or wants to see a recording, we'll be sending a recording around tomorrow by
00:00:48email for those watching on Big Marker. But before we kick off, let me just introduce myself and
00:00:53Career Foundry. I'm William, Events and Communications Lead here at Career Foundry.
00:00:57And Career Foundry is the online school for your career change into tech. So we guide you from
00:01:01complete beginner to job-ready professional in data analytics and help you land that dream job.
00:01:07With Career Foundry programs, you get a mentor and a tutor. So that's our dual mentorship model. And we
00:01:11also have a job guarantee. So if you don't land a job within 180 days of graduation, we give you
00:01:17a
00:01:17full refund. If you've got any questions about the curriculum or the program, I would say book a call
00:01:24with one by program advisors. If you're on YouTube, we've got a program advisor link in the description
00:01:30below. This evening, it's all going to be about trends. But if you want to ask anything specific
00:01:35about maybe the curriculum or the job guarantee, I would recommend booking a call with a program advisor.
00:01:42I think that's about it for me, Alex. I think I'm just going to give the floor to you. One
00:01:47thing I
00:01:48would say is that we are having a live Q&A at the end. And I'm sure we're going to
00:01:52have lots and lots
00:01:53of questions about emerging trends. So drop all of those questions in the public chat. Alex is probably
00:01:58going to pick up on some of those as we go through. But I'm going to disappear into the background,
00:02:03I'm going to come back at the end, and then we're going to have a live Q&A. So if
00:02:06you've got any
00:02:06questions about data analytics, now is the time. And Alex is the person because he has a fantastic
00:02:13YouTube channel, Alex the Analyst over on YouTube, which is closing in on 700,000 subscribers. So
00:02:20anyone from the Career Foundry audience this evening, check out Alex's channel and like and
00:02:25subscribe and help Alex get to 700,000. On that note, Alex, I'm going to disappear. I'll be back at
00:02:30the end. And yeah, everybody enjoy. And yeah, see you later. Awesome. Thanks for the intro.
00:02:37This topic I'm really excited about because 2024 is every year, just wild things happen in the data
00:02:47world. And so 2024 is no different. I think 2024 specifically has already been just wild in the
00:02:56past month. And so yeah, kind of forecasting out for the rest of the year. I'm excited to
00:03:01talk about it, kind of give you my thoughts. I think I have a pretty good perspective on this.
00:03:07My job, which now I own, I run a consulting company in data analytics. I work with a ton of
00:03:12startups. So I got a really good pulse and really good beat on kind of the startup world with a
00:03:19lot
00:03:19of data infrastructure companies, data migration companies, data analytics companies that I do
00:03:24consulting for. But then I have my YouTube channel and I'm on LinkedIn and Twitter. And so I get a
00:03:30lot
00:03:30of messages of people with new technologies that are coming out and questions and they're sending me
00:03:35articles and I keep up with the stuff a little too much. I probably should be doing other things
00:03:42sometimes, but I just get really into it. So I feel like I have a pretty good idea, a pretty
00:03:48good sense of
00:03:49what's kind of coming and some of the big trends that are going to be coming in 2024.
00:03:56If you do not know who I am, my name is Alex Freeberg. I am the founder of Analyst Builder
00:04:01for technical interviews and of my consulting company called Alex Analytics and then my YouTube
00:04:08channel. And so I do a lot of stuff. And if you haven't seen my YouTube channel, go check it
00:04:12out.
00:04:12It's pretty cool. Let's start off with, I specifically ordered these in a very intentional way.
00:04:21So this was not random. I want to get the big one out of the way because this one is
00:04:27something that
00:04:28I'm sure every single person out there is thinking about, talking about, always trying to keep up with.
00:04:35But, but AI specifically, there are some things that I think are going to be bigger than others.
00:04:43For example, I think everybody knows about like AI encoding and AI doing like summarization of text.
00:04:50So I think that is something that basically almost everybody's aware of at this point,
00:04:55especially if you're in the data world, like coding, you know, there's lots of different AI tools,
00:05:00tools like GitHub Copilot, ChatGPT, there's, you know, Bard and Anthropic and all these other
00:05:08different AI tools that you can use for coding. Some that are integrated, some of that are
00:05:12standalone products. So I'm not going to go super into that. Let's talk about a little bit about AI
00:05:17and kind of the changes that we're going to be seeing from 2023. 2023 was a wild, it was a
00:05:25wild ride.
00:05:25It was like a roller coaster with AI. When it first started kind of emerging in late 2022,
00:05:32there was a lot of, you know, I don't want to say hyping it up a little too much. But
00:05:40even into 2023,
00:05:41we saw a lot of hype around AI and whatever, everything that it could potentially do. There
00:05:47were a lot of fears around it, a lot of concerns about, you know, job, job layoffs because of that.
00:05:57And, you know, now we're into 2024. And I think people have a much better sense of what AI can
00:06:02do,
00:06:02right? It isn't going to replace software developers or programmers or data analysts,
00:06:07you know, anytime soon, but it is a fantastic tool. And now we're seeing AI can do a lot of
00:06:13different
00:06:13things. And what's amazing with AI is that now large companies are really starting to build out
00:06:19their infrastructure and build out everything they need to start using AI much better.
00:06:24For example, I attended a conference last year near mid-year, end of year. And this, you know,
00:06:31Fortune 500 company was talking about how they're going to be using AI across their entire enterprise.
00:06:38And that was the first time that I had seen it. But then after I started, you know,
00:06:44keeping up with some of these other companies, they also were talking about something called
00:06:47synthetic generated data, which is the third point on the thing down there. So I'm going to
00:06:51talk about that for a second. They mentioned something called synthetically generated data.
00:06:56And at first I was like, okay, what is this? How does this work? Because I've never heard of it
00:07:00before. Um, and essentially what it is, is it is data that you can generate based off of different
00:07:08use cases and different scenarios. For example, you have stock data, which most people are aware
00:07:14of. It's kind of something that a lot of people use in like projects and different stuff, or maybe
00:07:17it's cryptocurrency, uh, uh, you know, data. And so what you can do is you can plug in different
00:07:24what if scenarios. What if, you know, we have a Republican president nominated in 2024 versus a
00:07:29democratic president. And it does, it kind of forecasts based off of historical trends
00:07:35with that data. And it does these what if scenarios. So it forecasts, uh, and can create data
00:07:41for that scenario versus this scenario. And so it's pretty incredible and you can essentially get it to
00:07:49do almost whatever you want. Now, is the data perfect? No. Uh, but it is a really, really amazing
00:07:56tool to test out these different hypotheses and these different scenarios that could happen.
00:08:02Um, this is a really useful tool. I know I worked in the medical field, uh, in, uh, healthcare for
00:08:08a
00:08:09long time, working with cancer data, uh, hematology data, claims data. Um, you know, really, uh, and I,
00:08:16just for some context there, I've worked with a lot of clinical trials. And so with clinical trials,
00:08:21you have to get a lot, a lot, a lot of data, you have to collect a ton of data
00:08:25to make sure that
00:08:25your outcomes are good for certain drugs before you put them on the market. Well,
00:08:30with synthetic synthetically generated data, you might be able to help generate certain pieces of
00:08:35that without having to necessarily collect every single piece, which may take, you know, months extra.
00:08:41Um, and so there are a ton of real, really, really good use cases where you're using AI to,
00:08:47you know, add data or, uh, create this data for you. So really, really neat. Um, so if you haven't
00:08:54heard about that, highly recommend looking into it. It's, um, it really was pretty amazing when I
00:09:00first heard about it. Now I've seen lots of other companies, uh, start to do that as well. So I
00:09:04think
00:09:042024, uh, if you haven't already seen it, you'll see it more. Um, especially in like financial
00:09:11institutions, I think finance and healthcare, uh, are two where this specifically is going to be
00:09:17a really, really, uh, you know, just great thing that AI can do. Uh, the one up top is hyper
00:09:23personalization. Now, if you don't know what like personalization is for products, things, think ads,
00:09:29right? Uh, you go on the internet, you got a completely new laptop. You don't, you don't log into
00:09:34anything. Um, you create like a new, uh, LinkedIn account or Instagram account or whatever it is,
00:09:40and you start going on places and start going on websites and they start collecting that data and
00:09:45tracking that data. And they, on most of these websites are going to offer something like a
00:09:49personalized advertisement, right? You start looking at boats online and all of a sudden you're seeing
00:09:54boats, uh, in, you know, different things. We've already started to see these, uh, this personalization
00:10:03for the past several, several years, right? This is nothing new, especially with the, the invent of,
00:10:07uh, social media and the rise of social media. There's so much data to be collected. Well,
00:10:12hyper personalization goes a little bit beyond that. It goes a little bit beyond just collecting
00:10:17a little bit of data to where it's using your interests and it's using the knowledge that the
00:10:23data that they have on you to actually create a unique advertisement just for you. Um, now this is
00:10:29just an example. Hyper personalization goes, uh, beyond that, but for this example, for advertisements,
00:10:35you're going to start seeing that are unique to you, uh, something that no one else in the world
00:10:40is going to get the same advertisement as you. Um, I've already seen examples of this and I'm sure
00:10:46maybe you have as well on social media where, um, you know, I go on there, I'm Alex Freeberg.
00:10:52And so I go onto the internet and I'm typing around and it says, Hey Alex, you know, you live
00:10:57in
00:10:58Charleston, South Carolina. He's a coupon for 20% to this restaurant, uh, that you love, or that you
00:11:02went to that one time. It's kind of freaky. Um, but it's like super personalized and it's hard to
00:11:09kind of look away cause you're like, this is so in tune with who you are because they have so
00:11:13much
00:11:13data on you. Now imagine this for not just advertisements, but imagine this for a lot of
00:11:18different things. Well, we've already seen, uh, a little bit of this. If you think, uh, video games,
00:11:24if you think social media, um, and, uh, things like your iPhone, hyper personalization is going
00:11:31to be, uh, very, very, very big, especially probably in 2025, but we'll see a lot of it this
00:11:37year. Um, Apple is working on a lot of these things. And so, you know, do a little bit of
00:11:44Google search on the, uh, hyper personalization that Apple is doing. It's really fascinating.
00:11:48It's a little bit scary if I'm being honest. Um, because I see this like 10 years down the road,
00:11:53I'm like this, what, what, what's going to be happening this year will be, you know,
00:11:57what kids are growing up, uh, you know, in 10 years, that's going to be normal to them.
00:12:02It's just weird to think about. Um, and then the next thing is integration into everyday tools and
00:12:06products. Now, this one, I kind of mentioned a little earlier for coding, uh, just for one
00:12:11example is AI is getting integrated into almost all coding, uh, tools because everybody wants it.
00:12:17For example, I, for my company, Analyst Builder, we use Supabase for a lot of our data.
00:12:22Yeah. Well, uh, Supabase has an AI tool built in for questions and I'll say, okay, you know, what,
00:12:28uh, you know, let's think about a certain population. What are the people who are paying
00:12:32versus not paying? What, what are the percentages? And I can like, it'll help me generate a query.
00:12:37Does it always get it right? Uh, not always. Uh, anyone from Supabase is out there listening.
00:12:44Um, it's definitely not perfect. I've used it quite a bit and it's not, it's not perfect,
00:12:47but it is really good, right? It helps you get like 80% of the way and you can kind
00:12:51of
00:12:51fix it. Well, this is not just going to be for coding tools. Uh, you'll see this in almost
00:12:56everything. And I'm sure a lot of you guys, uh, are familiar with Google plugins. Google plugins
00:13:02are everywhere and AI is in a lot of them. And I've used a lot of them myself because I'm
00:13:07really
00:13:07interested in that for productivity. And, uh, I do a lot of, uh, summarizing information because
00:13:13I don't have time to read 20 pages of a PDF. And so we have AI that's being integrated into
00:13:19websites. You have it that is being integrated into, um, like PDFs and different tools. And so
00:13:26this is the year that you're going to see just AI everywhere, right? Um, and you can choose to not
00:13:32use it. You can choose to just do it the old fashioned way. And I say old fashioned, I just
00:13:38mean
00:13:38like how you do it now. AI is meant to be an enabler. Um, and it's going to be in
00:13:45almost
00:13:46everything. Uh, you're already seeing it on like a flight. Um, what's it called? Like
00:13:51Expedia.com, right? You go onto Expedia, you go onto a car website, they have AI, uh, they're
00:13:57trying it out. They're testing it out. And now that's in 2023 and 2024, it's going to get
00:14:02a lot better. There'll be a lot of, uh, products that are created specifically for car dealerships,
00:14:09specifically for airline, uh, websites, specifically for all these things. So you're getting the AI
00:14:15that's personalized. If you want to go back to the previous ones, specifically for each
00:14:19of these use cases. Um, and that's just a, it's a really, really interesting time that
00:14:27we're going to be living in, in 2024 to experience all this. Cause I guarantee you, cause I have three
00:14:31kids in five, 10 years there that's going to be normal for them. All of these tools that we're
00:14:36building and that are being developed and starting to be integrated into everyday products. I'm going
00:14:41to tell them about the time that I lived before AI, uh, before AI was in everything. Here's what
00:14:45it was like. And they're just going to be like, Oh, dad, you're so old. Um, let's go on to
00:14:51the next
00:14:51one. Um, this one is one that I've been saying for probably about a year, over a year now.
00:15:00And when I first started talking about it, I actually got a lot of pushback, um, because
00:15:09most people have a tough time believing or a tough time seeing themselves as entrepreneurs
00:15:15or as individual workers. Give me one sec. Um, I talked to a lot of people and a lot of
00:15:22people
00:15:22were like, I could never do that. I can never be a freelancer. I can never be a consultant. Um,
00:15:27um, because that my brain just doesn't work that way. And I need to work at a company where I
00:15:33just
00:15:33have a W2 and that's what I do. I started saying that maybe like the end of 2022, 20, beginning
00:15:39of
00:15:392023. And again, I got a lot of pushback, but now I've seen a big, a big shift in opinion
00:15:46on this.
00:15:47And I think it's becoming a lot more mainstream and people like, okay, freelancing is like,
00:15:50it's going to be pretty big. And I, myself, I feel like I had a pretty good pulse on that
00:15:56just
00:15:56because I started doing consulting myself, uh, about three years ago. And I was like,
00:16:01I was like, there's going to be a big market for this and even bigger market than there already was,
00:16:04um, you know, in a year or two, which is where we are now. Now what's leading, uh, what's contributing
00:16:12to this? There's a few different things. One is cost of living. Uh, the next is Gen X. And then
00:16:16the
00:16:17third is lifestyle. Now, why do I say these three cost of living is increasing. Um, people don't
00:16:23want to, uh, and I say people, I'm not talking about every single person in the world. I'm just
00:16:28talking about more people are starting to more lean this way because of these factors. Uh, but
00:16:34cost of living is going up. There's no doubt about it. Inflation really hit us all. Um, my everything
00:16:40bill for my whole life went up like 20% on the past, like two years, not good. So cost
00:16:46of living is
00:16:46really getting to everybody. Um, and because of that, people are looking for extra ways to make
00:16:53money apart from their full-time job. And so freelancing is definitely one of those ways.
00:16:58If you have a skill that you can freelance and put out there and be like, Hey, I'm really good
00:17:03at
00:17:03Tableau. I'm really good at cloud platforms. I really program well in R do this or that you can put
00:17:09that skill online and you can freelance for that skill. Um, consulting is a little bit different. I see
00:17:15freelancing and consulting. A lot of people will put those two together, uh, as a, as like they're
00:17:20the same thing. They can be in some instances, but consulting is more like you have a lot of
00:17:25experience and you're in there to advise, not typically do hands-on work. Um, not writing the
00:17:32code, building the infrastructure, building the visualizations, whatever it is. Typically that's
00:17:36more of like a freelancer. That's what I would, I usually, um, delineate those two by saying freelancers
00:17:42are more hands-on writing, um, as if you're like hiring an employee. Consultants are more like
00:17:47high level advising and bringing insights into a company. So, um, the next, the other piece of this
00:17:56was the startups. Um, so I've talked a little bit about freelancing and consulting, but startups are,
00:18:01um, popping up everywhere. I don't know if you've noticed, but more and more and more people are
00:18:08starting YouTube channels, TikTok channels, uh, Instagram channels. That is the example, because I
00:18:13think that's the most visible thing that people see. They're on social media. They see all these
00:18:17new, uh, uh, people coming up and starting their own thing and trying to start their like own little
00:18:22business. Well, that's the tip of the iceberg. That's what's visible. But all these people on the
00:18:28back end are also, a lot of these people are building products as well. They're starting companies
00:18:33to solve certain problems. And the reason for that, something that's driving that is, um, there has
00:18:40been quite a few layoffs in 2023. And now we just saw another wave of that in, in 20 beginning
00:18:46of
00:18:462024. Now, is that across all of, you know, the, uh, job market? No, but for some specific big tech
00:18:54companies, you know, that's kind of what people look at and they see that. And so all these really
00:18:59smart people who are in, were in these jobs are getting laid off. They're like, I don't want to go
00:19:04back. I don't want to go back to working at Amazon again, where I'm just going to get laid off
00:19:09in six
00:19:09months or a year. Cause the market is like just really wild. And so they're like, what can I do
00:19:15to not have that happen again? I'm going to go start my own startup. I want to solve this problem
00:19:20that I was working on in my previous company. Um, and I want to, you know, market that and make
00:19:25that
00:19:25a product so that I don't have to work again, right? And just automate it or something.
00:19:32Um, the next piece of the contributing factor is Gen X and lifestyle. Now, uh, I myself am a Gen
00:19:41Z.
00:19:41My, my wife likes to, uh, laugh at because, um, I'm right on the cusp, right? I'm like, I'm like,
00:19:47right, like two months before being a millennial. Um, and my wife is a millennial. And so, you know,
00:19:54our, the generations are just a little bit different. I think the older generations are
00:19:58more like, you know, company man, be a union man, and they want to like be with a company for
00:20:0230
00:20:03years. Well, uh, Gen X is not that way. I've already, uh, you know, worked with a lot of people
00:20:09in Gen X and they like to do their own thing. They are very independent. They don't really, uh, not
00:20:14everyone. I'm being very general, not everybody, but a lot of people there, you know, the, the way
00:20:19that the younger generation is going is they want to be more free. They want more freedom to move
00:20:24and travel and do these different things instead of being locked into a location. And these startups
00:20:28and freelancing and consulting, consulting jobs allow you to do that at any age. I'm just saying
00:20:35Gen X, uh, is definitely, you know, more leaning that way. If you'd look at, um, you know, just
00:20:40trends, the older generations tend to, you know, stick with the companies while the younger tends to
00:20:44be more like, I want to do my own thing. Uh, lifestyle is definitely changing as well. Uh,
00:20:50just to touch on this, uh, for a second on myself, I used to work in a, uh, uh, for
00:20:55a fortune 500
00:20:56company. I was in person and, you know, I got to meet a lot of really cool people, make great
00:21:01network.
00:21:02Uh, you know, it was really great. I liked it. Good perks, good benefits, but I didn't get to spend
00:21:08as
00:21:08much time with my family as I wanted. Um, and I have three kids now. And when I was doing
00:21:12that,
00:21:12my kids are very young. And so they were in daycare because my wife was working, um, at the
00:21:17same time as I was. And so we didn't get to see our kids that much. Um, not exactly the
00:21:22best lifestyle
00:21:22in the world. If you really love your children. And I really love my children. I wanted to be
00:21:26around them more. And so I started consulting on the side and eventually that became my full-time job.
00:21:31And so now I get to spend probably 40, 50% more time with my kids because I I'm there
00:21:37in the morning,
00:21:38uh, to help them get ready. I'm there in the afternoon when they come home, um, at three
00:21:42o'clock instead of like getting home at five 30, like I used to do. Um, and it's, it's awesome.
00:21:48It's really great. It's very freeing. Um, the next point on here is companies need to build
00:21:53data teams, but don't want to invest in entire team right away. This one actually, I will say kind
00:22:00of overlaps a little bit with the AI one. AI is something that a lot of companies, even if you're,
00:22:07they're like not even like a huge data company. Um, they, they do things more like the traditional
00:22:12way back in the day of intuition and, you know, uh, just knowing the market. Well, they're seeing
00:22:20like, Hey, our competitors are using AI and people like it and people are enjoying it. This new feature
00:22:25that they add in. We need to do that too. We need to not fall behind in times and be,
00:22:30you know,
00:22:3180 years old, you know, saying, this is how we do it. We're not changing our ways. We need to
00:22:35be
00:22:35adaptable. And so I've already worked with a lot of companies. Uh, probably last year I worked with
00:22:40maybe seven or eight companies that were not originally didn't already have huge data teams
00:22:47built out. That just wasn't their, like their core of their business. And there's more, it was more
00:22:52about sales and the product. That's what they were building, but the data wasn't like a huge factor for
00:22:57them. And so I've noticed that these companies are like, well, we have to, we don't really have a
00:23:02choice. We need to, but do a lot of these startups, especially don't have a budget for it. And so
00:23:08who
00:23:09are they going to hire if they don't have a budget to hire on, you know, three full-time people
00:23:13to
00:23:14build this out and start making it usable and have a business impact. We're going to do freelancers or
00:23:18consultants. Um, and a lot of freelancers who are really good at this stuff can do a lot of it
00:23:24on
00:23:24their own. And they're going to be a little bit more expensive, right? Cause freelancers and consultants
00:23:29typically charge a little bit more, but that one person can do a lot of this work and they don't
00:23:36have to pay for, uh, health benefits or insurance or all these other things. And so they're not,
00:23:41they don't have to commit as heavy into, um, hiring people on full-time. They can just hire these,
00:23:47uh, freelancers or consultants. And so I think you'll see, and before I go on the next one, I think
00:23:52you'll see as my last point here, you're going to see a lot more people, um, going into freelancing
00:23:58and consulting. Now, one note is it's hard to get into freelancing and consulting. If you have
00:24:03no experience, it just is, it's a fact. Um, you know, typically those, uh, the people who hire
00:24:09consultants and freelancers, they're hiring them to solve a problem, not to train them,
00:24:14right? Because they're not bringing them on as a full-time employee. They are hiring them for a
00:24:17specific reason for them to solve a problem. So you usually need experience for that just as a,
00:24:22as a side note, uh, this one, uh, this one has been really, really, really interesting to me
00:24:28because I've had some really close hands-on, uh, uh, experience with this just last year.
00:24:34I was working with a company and, um, I, at my previous, one of my previous roles, uh,
00:24:40the fortune 500 company, I was working on data migration, um, up, uh, getting everything into
00:24:46the cloud because everything was all spread out. We had all these data silos. So I was working on data
00:24:50migration and, um, you know, choosing a, a data cloud platform and, you know, contracts and all
00:24:56these different things. And so I have really good experience with that. And so, you know, this,
00:25:00this company was like, Hey Alex, um, you know, we want to know, should our company move everything
00:25:07into the cloud or is there a different direction? And so I started talking in, in, you know, there may
00:25:13have been someone who's better at this than me out there, but I think it was some of the most
00:25:19interesting conversations that I've had around this. And I, I, you know, it was very helpful,
00:25:23but here's the point. When we started really getting down to it, I said, what's your main,
00:25:29like, what is your goal with this? They said, we want all of our data in one place. And I
00:25:32was like,
00:25:32totally get it. I absolutely understand. And then they said, security is like our number one thing.
00:25:37You know, we've had, um, we, we've just heard horror stories of data security and all these
00:25:45different things. So I started asking more and more and more questions. And well, by the end of
00:25:49it, I was like, you know, I'm a big believer in the cloud. I'm a big believer. And as are
00:25:56a ton of
00:25:56other companies, I was like, but for your use case, I don't know if I'd recommend the cloud for you
00:26:02guys
00:26:02because we looked at their budget and we looked at their trends for, or, you know, their projected
00:26:07revenue. I was like, I think I would just bite the bullet and, you know, do everything on-prem
00:26:13them because it wasn't a huge company. And I was like, you're going to have exorbitant fees
00:26:19for your, to build out your infrastructure and then also maintain it. And we were looking at like
00:26:24AWS. I was like, that's going to be a lot of money, a lot of money. And I was like,
00:26:29for what
00:26:29you guys are doing, I don't think that's what you need. I was like, I, if it was, if this
00:26:32was my
00:26:33business and I'm trying to help you run this as if it was my business too, I was like, I
00:26:37would just
00:26:38bite the bullet, pay the upfront costs of building an on-prem server or servers for them. And that
00:26:45will by far be an incredible investment, have better security, less downtime, less dependency
00:26:52on a cloud platform. And some of these things that I have right here are the exact reasons why I
00:26:59recommended that for them. So let's talk a little bit about why it's called the great cloud divide.
00:27:05Right now, there has been a huge uptick over the past 10 years with cloud platforms, just
00:27:10massive, multi, multi, multi-billion dollar revenue streams for all these cloud platforms,
00:27:17specifically the ones that most people know is Google Cloud Platform, AWS, and Azure. Just massive,
00:27:23right? And there's a lot of reasons why people want to go to the cloud. It's easy. It is genuinely
00:27:31pretty easy to get up and to use these platforms in general. You get to put all of your data
00:27:38in one
00:27:39place, maybe even easier than doing something like an on-prem server. And I would say that when you're
00:27:47kind of spread out, when you're a larger global company, it's easier to connect to all these
00:27:52different places and give access to the data, which is very important. Now, the downside of cloud
00:27:59platforms is there is a massive amount of cost associated with cloud platforms, and it's only
00:28:06going to go up. It's only going to get more expensive. There's no, there is no possibility
00:28:13in my mind that we will see a decrease in prices for cloud platforms in the next five years. It's
00:28:18just not, it's never going to happen. And so these cloud platforms there, they typically will have a lot
00:28:25of incentives for you to get into their ecosystem. But then once you get in, you're going to have a
00:28:31ton, a ton, a ton of fees and costs associated with that. So just to give you an example, if
00:28:37you've
00:28:37never seen like these memes before, it's usually a guy who's like, you know, opening an envelope and
00:28:44he looks at it. It's like, this is my AWS bill. And his eyes get huge and he's shocked. And
00:28:49he's like,
00:28:49like, this is insane. It's because AWS is not like, just as an example, AWS is typically you
00:28:57have something like an autoscaler where you're scaling certain features and you can't always
00:29:01control what you're going to pay. Right. And so for a lot of these companies, they turn on these
00:29:05autoscalers and like, Hey, you know, if it costs 500 bucks, no problem. It's 500 bucks, 500 bucks,
00:29:10500 bucks. Then one month, something changes. Whether AWS tells you or not, or you're, you're where
00:29:16you're, you know, ingesting the data or whatever changes. And then the next bill is like $3,000
00:29:21or $6,000 or $10,000. I've heard instances where a bill normally was $10,000 went up to a
00:29:27hundred
00:29:28thousand dollars. That is very, very scary. And so yes, usually they have incentives to get you
00:29:36into their ecosystem, but the long-term operational costs for these cloud platforms are very high,
00:29:40which is okay. If you're like a fortune 500 company, a big company, um, there are good plans,
00:29:45even for small startups, a lot of startups use, uh, AWS and Azure and all these, uh, different
00:29:51cloud platforms, but long-term it just, it never, it always goes up. It always goes up. So because of
00:29:58this, because of that piece of it, there is this growing movement towards more on-prem servers for
00:30:06people's data and on-prem servers are just their servers, right? So like in the cloud, they have all
00:30:12these server farms where it's literally tens of thousands of servers running at the same time.
00:30:18And when you use their services, they allocate you a specific server that you're just kind of like
00:30:22borrowing. You're just like renting out a server essentially. Um, with buying a server, you're,
00:30:28you're, it's like taking one of those servers or multiple of those servers and you're putting it
00:30:32on-prem, which means on-premises you're, you're buying a physical server that you get to control.
00:30:36And it's a little bit more upfront cost. Uh, typically you'll need someone who can work it
00:30:43and run it. So that typically is, uh, like a single, could be a single person or an IT team,
00:30:48um, who actually runs and maintains the server, make sure everything's, you know, good in the
00:30:53databases and, and with the data ingestion and the servers running well, and it's a more upfront
00:30:58cost a hundred percent. But if you kind of look further ahead, a lot of these companies are
00:31:06looking further ahead and they're like, our AWS bill is going to be 75,000 per month if we continue
00:31:12down this road. But if we spend $250,000 now, it's going to only cost us like 10,000 a
00:31:19month for the
00:31:20next 10 years. Or, you know, that's like a rough example, but I've had those exact conversations with
00:31:26many companies and it, it is this growing, a growing movement of, yes, the cloud is still growing,
00:31:33but you're now seeing a lot of people just bite the bullet and be like, I can't, we can't afford
00:31:40the, our AWS bill for the next five years. Like it's just not feasible. So, um, that is, that is
00:31:46definitely a big reason is just cost. The next things are internet dependency and downtime. Um, for
00:31:53cloud platforms, you are dependent on their services. If their services are down, even for 10
00:31:59minutes, sometimes certain companies cannot, they cannot operate effectively for their product or
00:32:06their customers with any downtime. Now, downtime can happen with your servers that you buy on-prem,
00:32:13right? But most likely you're going to be in control of when that gets back up. You have somebody
00:32:17on-call or you have somebody in the office who can fix that quickly. You are not in control when
00:32:24it is
00:32:25in the cloud. You are fully dependent on them. And that is a problem for a lot of businesses.
00:32:29They cannot afford to have that. And they've been dependent on it. And they're like, we can't do that
00:32:33anymore. Um, some companies aren't like my company. We're not, uh, we are reliant on AWS being up,
00:32:40but if it goes up or if it goes down, it's, uh, you know, we have backup redundancies in place.
00:32:47And so these are all things that you need to be thinking about with, you know, these two different
00:32:51options. The last one is data security. Um, most people would agree. And I would say most,
00:32:58I'm going to say all that on-prem servers are going to be more secure than, uh, cloud platforms.
00:33:03Um, your data isn't sitting on some server where someone else can go access it. Your data is with
00:33:09you on-prem, uh, you know, controlled by some type of database administrator or something like that.
00:33:15Much, much, much more secure typically in general. And so the cloud, uh, is going to grow, but we're
00:33:22seeing a larger move towards on-prem. Um, the next one is the year of documentation and governance.
00:33:29I don't know why I named it that. I just thought it'd be fun.
00:33:31Um, this one relates, um, a lot to AI as well. Um, all of these, you know, there's a little
00:33:39AI
00:33:39and everything these days, 2020, you know, what can you do? Um, people are realizing, and I remember
00:33:46back when I first started about seven years ago, documentation wasn't like that super important,
00:33:54especially like some of the smaller companies I first started working for. I was working at a
00:33:58nonprofit, not a healthcare and electric company. Uh, when I first started out, documentation wasn't
00:34:02like super important. Um, but with the rise of AI teams are like, wait a second, our data is
00:34:11important. Yeah. We realized that, but now AI can do more than just work with our data. AI can have
00:34:17context and it can have processes and procedures. And if we actually wrote that out and had, um,
00:34:23some guidelines and we could put that into AI, it would just help with it understanding our data
00:34:28even more. And so, um, I know multiple companies, uh, that are, you know, uh, just off the top of
00:34:37my
00:34:37head, excuse me, Conva and Figma are two really big, uh, tech companies. Both of them came out with,
00:34:45uh, you know, I don't, I don't, I don't know what the word is. Um, not newsletters, just like
00:34:52announcements being like, we need, we're, we're, we're doubling down on, uh, documentation and
00:34:58governance for those specific reasons. Hey, we're introducing all these new features and these new
00:35:02things, but we need to get our ducks in a row for it to be really effective and useful. And
00:35:07that's,
00:35:07there's, that's just two examples. There's a ton of other examples. Um, uh, and so what I wrote
00:35:13here was that government, uh, governments and companies are cracking down on their data
00:35:17governments and compliance because historically it just hasn't been crazy important. Um, and it is
00:35:23shocking if you really think about it, because just with the documentation piece, what used to happen
00:35:31is, is there was two or three people who had it all in their head. Well, two of them would
00:35:35leave.
00:35:35There's only one. So the one guy would train the other people just through training, right? Hands-on
00:35:40training. There was no documentation at all, like just none. Um, and those days are still here,
00:35:45right? They're not going away, but you're seeing a big wave of people who are realizing for their
00:35:50use cases for, especially when using it with AI that they really, really need that. Now the next
00:35:55piece is, uh, uh, data governance and compliance and data governance and compliance are usually refers
00:36:02to, uh, a lot around how you're actually protecting your data, the data quality, um, access
00:36:10to your data. So there's a lot of different, uh, uh, avenues and pieces of that equation, but
00:36:16data, this has been said since, you know, uh, maybe like 2010 or in that era of social media
00:36:24coming over that data is the new data is the new oil data is the new goal because it is
00:36:30so
00:36:31incredibly important. Um, with everything that's happening in AI, with all the security or security
00:36:38breaches and leaks and data, um, all the issues that people are having with AI trying to use their
00:36:44data, they're realizing that their data is one of their most important assets. And so they're really
00:36:49cracking down on security compliance. They're really trying to utilize that as best as they possibly can,
00:36:56because I mean, if you just take a step back and you look at what's driving your business for a
00:37:04lot of
00:37:04these companies, it's the data and they need to get it under control. And 2023, we saw a big surge
00:37:11in this. This year is a year where you're going to see a lot more companies making announcements. Hey,
00:37:17we, we did this initiative to secure our data. We did this initiative to make sure that these things
00:37:22are done. Um, and they'll tell clients because clients want to know if we're using your AI products
00:37:26that your data is good on the backend. Um, and so that's kind of like a, uh, it's a reassurance
00:37:31thing, right? And it's only going to be more important in years to come. Uh, and the last thing
00:37:36I'm just going to read this is documentation allows for better compliance, data quality management
00:37:40and risk reduction. Kind of some of the things I said. Oh, I'm talking too much. I'm already 37
00:37:47minutes in. I'm gonna have to go over. Well, I hope you don't mind. Well, I'll get to, I'll, I'll
00:37:53try
00:37:54to not talk too long and then we'll have time for Q and A. We'll do that. All right. Data
00:37:59unification.
00:38:00Um, data unification is really the, and I've mentioned a few times, but the idea that you can
00:38:07really use all your data. Now, uh, I read this, I think on like a medium article. So don't quote
00:38:13me
00:38:13on this. I don't know if this is a hundred percent true, but it, it basically said, um, the average
00:38:18company uses about 30% of their data. And the reason is not because the 70% of their data
00:38:25is
00:38:25bad. The reason is because all their data is so spread out. Only certain people have access to
00:38:30certain pieces of the data in Excel files, um, in different databases. And it just can't be all
00:38:36used, which is really, you know, if we're talking just about before how data is the new oil data is
00:38:42gold, then you're only using 30% of it. That's not good. That's not a good thing. Um,
00:38:48and so, uh, data silos are kind of this term that most people have heard of where it's
00:38:54like, I have my data in a silo over here. I also have data in a silo over here. We
00:38:59want
00:38:59to use both of them, but because of how we've set up our systems, uh, this is in some weird
00:39:05database over here. These are just in PDF files and Excel files and, uh, you know, some one
00:39:10drive over here and we can't use them. We can't connect it and use it even though we really
00:39:15want to, or if we do, we have to export it into like Excels, merge it together, uh, with
00:39:22a manual process, upload it to this SFTP, which then automatically ingest it into the
00:39:28system. I'm not speaking. I actually, let me take that back. I am speaking from personal
00:39:32experience. I was going to be, I was going to try to say something funny. I, that is absolutely
00:39:36what I used to do or have to do with some previous companies before we set up some more
00:39:40automation, uh, in our processes. That's how a lot of companies are. Um, and just as all
00:39:47these other things, data is becoming so incredibly important. People are like, we can't sit around
00:39:51and not use all this data that could be beneficial in this area or this piece of our product or
00:39:56this for our customer service or this or this, we need to bring it together and we have to
00:40:01do this. Now there's a lot of different avenues for this. Um, but one that I think a lot of
00:40:05people
00:40:05have heard about or, or, you know, looked at, or something like Microsoft fabric, which is
00:40:11meant to literally put all of your data into one place. And it's meant to keep it extremely
00:40:16organized. And of course it's under the Microsoft umbrella, which has a really great, uh, uh,
00:40:21brand recognition. And, um, you know, that is exactly why they designed this. Now there are
00:40:28other products and tools that you can do this that may have already come before that, but this
00:40:32is Microsoft's version of this. And they have something called one, like, um, I think I wrote
00:40:38it, uh, wrote what it does at the bottom. It says this one data lake, which a data lake is,
00:40:43uh, was a huge buzzword about like seven, eight years ago. Um, data lake was a huge buzzword. It's
00:40:50like, Hey, you can just put all your data in here and you're gonna have access to all of it.
00:40:54It's
00:40:54amazing. And then people realized they were like, wait a second, we don't know how to actually put
00:40:59all of our data in one place and create the correct schemas and, and, um, data architecture
00:41:04to actually use that data. That's our fault. Um, we're not going to use data lakes anymore.
00:41:09So it kind of like, it was a huge thing, like a huge movement. And then it really fizzled out,
00:41:15but it's still a great concept. And that's what Microsoft fabric was trying, is trying to do.
00:41:19And so the one data lake, uh, is meant to be for the entire company, entire organization to have
00:41:24one cup, one copy of data from, uh, uh, multiple analytics engines running at the same time.
00:41:31So you don't have to, um, you don't have to get it from this database and pull it in and
00:41:37get it from
00:41:37this database, pull it in, merge it, put it into this data lake, and then do all this. It, it
00:41:42allows
00:41:42you to access the data where it is, pull it in, read it, query it, use it all in one
00:41:48place. That's the
00:41:49idea. These types of systems are going to be a big hit, uh, with a lot of organizations,
00:41:56I would say specifically small and medium size. Initially, the large companies are always going
00:42:01to have an issue with this because you have years of data trauma, uh, and infrastructure trauma and
00:42:07architecture trauma that is going to take a long time to unravel. So, uh, larger companies will have
00:42:12a harder time with this, but, uh, small and medium size companies should be able to utilize these very
00:42:18quickly. Um, and it's just, it is something that I wish I had known about or what was kind of
00:42:24around
00:42:24back when, uh, I was in analytics and as a manager, because these are the exact issues that were really
00:42:33difficult to solve. They were really, really, really difficult to solve. How do you get this
00:42:37data over here and this data over here to connect? It's a universal problem. Um, and so 2024, you're
00:42:43seeing new tools coming out. You're seeing people really working on this because again, data is just,
00:42:49incredibly important. All right. This is my wishlist. This is what I want to see in 2024. Um,
00:43:01the first is, uh, I want hyper-realistic virtual worlds where I can do my business meetings in,
00:43:08because right now I have to do it like this where I'm doing like a webcam. It's not ideal. Uh,
00:43:14it's not
00:43:14bad. I don't mind it, but I do have a VR headset. Uh, it's not here, but I do have
00:43:18a VR headset.
00:43:19I mean, I'm over here in the metal world. I'm living it up, but it's just not good enough. Uh,
00:43:25it's just not good enough for me. I want something that's hyper-realistic and for other people to be
00:43:29able to access and do it as well. That is what I want. The next thing I want is a
00:43:33robotic, super
00:43:34intelligent assistant that helps me with essentially living my life. I want it to fold my laundry. And then I
00:43:40also want it to respond to emails. That is the dream. And that is what I wish for 2024. And
00:43:46the last one
00:43:46is real-time brain decoding for instant analysis, AKA what Neuralink is doing. I just need to sync up
00:43:52with Elon. Um, and generally I think I could boost my productivity by like 200%. I think if, if I
00:44:01had
00:44:01something like that, I would be the most efficient person in the world. That is our future. I just want
00:44:07it to happen now. So I think that's more than reasonable. I think that's more than fair. Um,
00:44:12of course I'm being a little bit, um, facetious or hyperbolic or whatever, uh, term you'd like to
00:44:18use, but a guy can dream. Now that is all I have. Uh, and I think Will's going to come
00:44:25back and we're
00:44:26going to do some Q and A. I went a little fast at the end. I could talk about these
00:44:30topics for each
00:44:31one of these topics. I could probably talk at like an hour on. Uh, and so I try to, I
00:44:36try to limit
00:44:36myself. Alex, I think we can see the passion shining through. I love it when you go do the
00:44:43deep dive, but I think it's great to get completely lost in the topic with everything that you said
00:44:49and all of the trends that you've mentioned, um, is now the right time to start off as a data
00:44:55analyst?
00:44:57Um, you know, there's, there's good and bad in what's going on right now. I would say the bad is
00:45:03we're seeing more layoffs. The economy's a little down, uh, uh, definitely not. It's not great just
00:45:12with inflation and everything and the layoffs. That's not great. Um, but that's, uh, unfortunately,
00:45:19unfortunately, that's just kind of how things go every so often. This is a site. This is a cyclic
00:45:23thing that we, if you look back, this is not new to 2023 or 2024, unfortunately. Um,
00:45:30so there is, we're kind of in like this little dip right now. I do think things are going to
00:45:35go back
00:45:35up. I, I, if you just historically, that's just how things go, unfortunately in terms of, is it
00:45:41actually a viable career for the future? A hundred percent. I think you're going to see, um, a lot of
00:45:47people joining and coming into, into the data world, um, for the long foreseeable future. I, I absolutely
00:45:56don't see AI replacing, uh, people on mass. Like we used to worry about back in early 2023. I think
00:46:04we really understand its limitations and understand what it can and can't do right now. And even with
00:46:08improvements, um, even with a lot of improvements, I, I was listening to, um, Sam Altman, who's the CEO
00:46:15of, uh, open AI. He was like, you know, everyone freaked out and panicked, uh, in 2023. And now all
00:46:23they're doing is complaining about how slow it is and how it doesn't give good outputs and how it's
00:46:26terrible. He's like people after a couple months, um, you know, it's just become like part of their
00:46:32tool chest. And he's like, it hasn't replaced anybody really. And so he's like, um, you know,
00:46:38I, he was specifically talking about AGI. He's like, you know, one of these days in five, 10 years,
00:46:44we'll hit AGI. It'll be mass panic for two weeks. And then after that, people will go on with
00:46:47their lives and do what they've always done. Um, and so I, I, I'm not as like on that side,
00:46:53I definitely think it, it, it will play an impact, but I absolutely see, uh, there being more jobs
00:46:59created with AI and more teams and companies opening up their doors to data professionals
00:47:04than have ever before. And so I personally, I think we're just in a little bit of a downtrend.
00:47:09Um, and you know, that's, it is unfortunate. I hate this. Like, I don't like seeing when my friends and,
00:47:15and people who I know lose their jobs, uh, uh, and I have recently, it's, it's not fun to see,
00:47:21but there is going to be an upswing and there is going to be a, a big drive and push
00:47:28back into the,
00:47:28you know, data world in not too long, I think.
00:47:33Awesome. Thanks for that. Um, if anyone out there has got any questions on YouTube or LinkedIn,
00:47:37now's the time to ask also on big market in the Q and a tab, um, do drop them in
00:47:42right now and I will
00:47:43go through them. Um, I know this evening we've got a lot of people who are watching who are thinking
00:47:47about maybe transitioning into the field of data analytics. How do you think given these trends
00:47:53in AI, given trends in the job market, what do you think, uh, positions like for the beginner,
00:47:58you know, the entry level positions, are companies looking for, um, different skills maybe? Are they,
00:48:04are they honing in more on soft skills or cross-functional roles? Are you seeing changes
00:48:08in the market at that end? I, I, I, I haven't seen them myself. I just anecdotal anecdotally
00:48:16hear them here and there. I get a lot of people reaching out to me, um, who are like, Hey,
00:48:20you
00:48:21helped me get a job. Here's how I got it. And then I, and I read their story and it
00:48:24usually not always,
00:48:26but I've heard a lot of stories like this, which is, Hey Alex, I learned a lot of skills from
00:48:30you,
00:48:31but I used to be a nurse and I use my nurse, uh, my nursing background to get a job
00:48:38at a healthcare
00:48:38and analytics company, almost very similar to me. I see a lot of that for a lot of different domains.
00:48:43And so I think if, if I were to kind of read into that, which I, you know, again, I
00:48:50don't have like
00:48:51a ton of hands-on experience with that specific thing. I would just say that I think domain knowledge
00:48:56is going to become more and more and more important. Whereas maybe seven, eight years ago, uh, it was
00:49:03a lot on the skills, but now with, um, you know, skills being more widely available to learn and AI
00:49:13being able to help with the coding pieces a little bit here and there, and, uh, uh, AI being integrated
00:49:18into things. It's more about truly understanding what the data is telling you, what it means.
00:49:25And if you have experience like domain experience in a certain area and you're able to transition
00:49:30that into analytics, that is like, that's going to be like a golden ticket. Um, and that's why I've,
00:49:36I've, when I had about a year ago, I stopped doing about a year ago. I, for two years, I
00:49:40did mentorship.
00:49:41The people who I saw who were the most successful with transitioning careers were the people who had
00:49:46experience in a field already. So they were, you know, one was a lawyer, one was, um, a warehouse
00:49:53worker. So he worked, he understand logistics pretty well because of the kind of work he did.
00:49:57One was a teacher. One was a nurse. Every single one of them were using their background in some way
00:50:01to transition into, uh, the career into analytics. And I truly believe that that is only going to
00:50:08become more important in the future.
00:50:12A fantastic answer. I would say it mirrors exactly the experience than the graduates that we have
00:50:17through career foundry. Those with different backgrounds, transitioning into a new field.
00:50:21A lot of people think that they're going to be starting out scratch with zero skills,
00:50:24but actually you're bringing a wealth of experience from your previous position
00:50:27into your new position and, uh, you will see things with different perspectives. So great answer there.
00:50:32I'm diving into big marker. I see Joanna's got a great question specifically about AI.
00:50:37How trustworthy is AI and AI tools? Can we always fully rely on it?
00:50:44No, and I would not. Um, I, I've been using,
00:50:48ever since chat GPT really became a thing, like a thing that most people were using back in like
00:50:53late 2022, I've been using it very consistently. I used it today. Um, and, uh, well, and I've tried
00:51:01different variations. I've used Bard and Anthropic and, uh, GitHub co-pilots and different AI
00:51:07systems and tools here and there. Here's what I'll say is they're only so intelligent to a point
00:51:14right now. Now, could they get better in the future? I absolutely think they will.
00:51:18But even right now, I have so many issues with using AI almost every time I use it. It never
00:51:27gives me what I want. You have to really, you really do learn how to talk to it to where
00:51:31it
00:51:32understands what you're trying to do. But even then, even if you give it all the context in the
00:51:36world, all the information in the world, it only knows what it knows. It's not like this all
00:51:40knowing genius that just is like Einstein level in every single topic. It's just smart in every
00:51:46single topic and has a wealth of data and information to pull from. But even just looking
00:51:51at coding, I have gotten, I've gotten more frustrated at using AI for coding than I have
00:51:59been just doing it myself. Because in my head, my perception is, is AI should be able to do this
00:52:05because it's not that hard because I've done this before. And I could have written it myself and like,
00:52:1030 minutes or an hour. And I spent an hour trying to get AI to do 90% of it.
00:52:15And it got me like 60%
00:52:17of the way there. And I'm just super frustrated. So I ended up writing it myself. So just for coding,
00:52:22that's an example with coding, but you'll see that with a ton of stuff. For example,
00:52:27text summarization. I use text summarization a lot. I have a tool on my browser that when I'm on a
00:52:34website, it'll, it'll generalize it and I like it and it gets the bullet points. But then when I'm like
00:52:40skimming through the article, I'm like, wait a second, that's like super important. And it
00:52:43didn't even mention that in the summary. And I get, I get mad at the AI. I'm like, hey, AI,
00:52:48I'm like, why didn't you include this is like the most important part. And they're like, oh yeah,
00:52:52no, I should have included that. Sorry about that. I'm like, it, the more you use it,
00:52:57that truly, the more you use it, the more you realize its limitations. And it has a lot,
00:53:03especially with as tools like these, these tools become a lot more commercialized.
00:53:11They're getting almost a little bit dumber because they have to limit themselves for liability
00:53:17purposes. So like ChatGPT, you've probably seen this on Twitter or X or LinkedIn or wherever.
00:53:22They're like, hey, why is ChatGPT giving me answers like this now? Because this is worse than
00:53:28how it used to give me. They have to do that for legal purposes. And so I only think, I
00:53:35genuinely
00:53:35think you're going to get in the future, better answers, but those limitations are only going to
00:53:39get worse because right, I keep up with a lot of AI stuff right now. They're, they're the UK,
00:53:47a lot of European countries, America, Canada are working heavily on restricting AI models and how
00:53:54they're used and all these different things, which whether you like it or not is going to
00:53:58change how they work, the responses they're allowed to give, things they can do legally.
00:54:05And so, you know, I, I, I, there's just so many different aspects to AI that it's hard to get
00:54:12into
00:54:12all of them. I just don't see AI ever being like this perfect one tool that does everything.
00:54:17It, I just, at least not, not now, maybe in like, you know, 10 years we'll pop, someone will pop
00:54:23up
00:54:23and I'll be like, I was like, that really could work. That really could be the one tool for
00:54:27everything. But now, absolutely not, not even close.
00:54:32I think linking onto the AI track, Rita's also got a great question. Considering the pervasiveness
00:54:37of AI, what type of uses should someone trying to enter the market as a data analyst should practice
00:54:44or, or get, um, experience with related to AI? Yeah. That's a question that I've, um, had with a
00:54:52question that I've answered with a lot of different people who have messaged me who on this topic,
00:54:56they're like, Hey, is it even worth getting into the data field with AI? And every single time I'm
00:55:01like, yes, absolutely. Um, how I recommend like learning about it or using it is for usually two
00:55:09or three specific things. Uh, the first one is just coding. I think it's just really lends itself
00:55:15well to help you learn how to code. Um, not it, I've tried to get it to like teach me
00:55:21coding,
00:55:22but it only goes so in depth and you have to kind of know, already know how to code for
00:55:26it to really
00:55:26teach it to you well. And so it's this juxtaposition there, but it can help you understand a lot of
00:55:31concepts and give you examples, which is really great. Um, but in, if you're trying to use it,
00:55:38like how you would use that as a data analyst for coding, you know, you'll say, um, here's an
00:55:42example of some sample data. Um, how would I get this in my SQL? How would you write this question
00:55:50in my SQL? And then it'll write out the code for you. You can test it in your, my SQL
00:55:54database and
00:55:55you can, uh, uh, try to see if it works. If it doesn't, you can go back to AI and
00:56:00revise it or you
00:56:01revise it yourself. So that's just one way of coding. Uh, the next thing is I would say is it's
00:56:06really
00:56:07good with brainstorming ideas. Um, for data analysts in the real world, things are very,
00:56:15very nuanced and very, um, it's not black and white. There's a lot of gray. And so oftentimes
00:56:24you'll, you'll have this scenario where you're like, when you're learning just on yourself by
00:56:29yourself or, or on like a platform, it kind of is, you know, it's catered to help you learn,
00:56:35but real world scenarios are very different where you're like, is it best to do this? Or is it best
00:56:41to do this? Because those are two very different things, which will change how I get the data,
00:56:46how I clean the data, how I do all these things, which is best. Well, posing these things into AI,
00:56:52I found to be very helpful with brainstorming ideas because I'm like, you know, there are pros
00:56:57and cons of both. Tell me the pros and cons of doing it this way versus this way. Um, or
00:57:02tell me what
00:57:03other options I may have besides these two options. And it really helps me like, think about
00:57:07how I'm going to solve a problem. And so, um, there's more examples. Cause I, I don't want to
00:57:12talk on that question forever. Cause I can just keep going, but just those two examples, genuinely,
00:57:17you can practice that today. Um, and you can start working through that, get a data set from Kaggle,
00:57:24put it into a database, put it into a Python data frame, put it into R, put it anywhere you
00:57:29want
00:57:29and try to solve it using AI, try to ask it questions. Um, try to have it do things for
00:57:35you. You will see the limitations quickly. Um, the, the more you play around with it,
00:57:40the more limitations you see. And, um, it's a good thing. I think it's a very, very good thing
00:57:45for newcomers to see the limitations that AI has because most newcomers are like, AI can do this
00:57:52and I don't have to, um, I won't be able to get a job, but when they actually start messing
00:57:56around with it and trying it, they're like, um, yeah, no, I get what Alex is saying. Like
00:58:00this, this, they can't really do this piece or this piece or that piece. And, um, so it's just,
00:58:06it's good to, it's good to get that hands-on experience with it.
00:58:12Awesome. Thank you. And, uh, just to put a shameless plug out there, we have recently updated
00:58:16all the career foundries program with AI. So we are AI enhanced, especially on the data analytics
00:58:21program. Shout out to a career foundries curriculum team who've worked very hard on that, um, in the
00:58:25background. I'm going to switch over to YouTube. Uh, hi everyone over on YouTube, Alex's audience.
00:58:31There's a great question come in, uh, from a man. Um, I've been working for three years as a data
00:58:36analyst. What makes the difference between being a data analyst and a senior data analyst? So when
00:58:42we're looking at career levels. Sure. Yeah. I, um, I was never a senior data analyst. Uh, just so
00:58:50everyone knows I never was, I jumped straight from a mid-level data analyst and do a manager role.
00:58:54It was very odd. Um, so I never got to fully experience that, uh, senior title. Although I
00:59:01think I, I definitely was at the level of senior. I just didn't get that pay jump in those two
00:59:09years
00:59:09that I did. And I just went to be a manager, but here's what I will say when you're a
00:59:13beginner,
00:59:13when you first start out, um, there aren't a ton of expectations. Uh, you get into the job and it's
00:59:20like, you know, Hey, run these queries for us, debug this, um, you know, make sure this process
00:59:26is working. Right. And you just kind of, you learn, you're there to learn, you're there to grow.
00:59:31And the company is hoping that you'll stay with them a long time. And, um, you know, it's a good
00:59:35ROI. Well, when you become mid-level, the expectation is, is okay, you're going to take
00:59:41on some more serious projects. You're going to take on, um, more important work. And we shouldn't
00:59:47have to baby you as much. You should be able to do most of this independently. Like that is
00:59:52definitely an expectation. You should be able to do this independently. Maybe you're even kind of
00:59:56somewhat mentoring a little bit, like the new people, um, in your company. So if a new person
01:00:00is hired on as an entry level that you're there to kind of like support them and help them. Um,
01:00:06then you get to senior and the senior level is you are expected to work on your, your, um,
01:00:16the work that you're doing is expected to be a very high level of difficulty, um, as well as
01:00:25collaborating with a lot of other teams in order to make it work. For example, um, you know, back
01:00:32just when I was a mid-level, but again, I, I, I definitely was, I would say I was more
01:00:36like senior
01:00:36level, even though I didn't have that title, I was taking on big, big, big projects that had,
01:00:41you know, million dollar budgets. And, you know, we'd be working with three or four different
01:00:47departments on these projects and all of us are working collaboration and it's really advanced and
01:00:52really, really tough data pipelines that we're trying to create. And, uh, and we're working with,
01:00:57uh, just, it's very, very difficult, right? You can't, you couldn't give those types of projects
01:01:03to mid-level or entry level. It would just be too complex. And so the, the technical skills have to
01:01:09be much higher for senior level. Usually the domain knowledge is also much higher, um, because
01:01:15you have to know the data inside and out. That's back when I was working with hematology, oncology
01:01:21data. That was my specialty. Like I knew that data inside and out. I knew it extremely well,
01:01:26how it originated, how it got into our systems, how we used it for all of our products and all
01:01:32of our
01:01:32different things. And so it's just, it's a knowledge thing with both skill and with domain.
01:01:39And then you really should be working incredibly independently and helping mid-level and entry
01:01:45level. That is, that is the difference. And so you're just working on a lot more important work,
01:01:50um, that has a bigger impact on the business because they know that you can handle it and they
01:01:54trust you. Whereas mid-level, especially entry level, they're not going to give you those big
01:01:59projects that are worth a lot for the company. Cause you know, if you mess up or you don't know
01:02:04what you're doing, or you haven't had experience with it, you know, there's no, there's the lack
01:02:08of trust. And so senior level, they should trust you know that you're going to be doing a really
01:02:13good job and you know what you're talking about. Awesome. Great. Thanks for the answer. Um, also a
01:02:20couple of questions that come through about portfolios. I've seen the question by, uh, Bianca on
01:02:24Big Marker. Um, and just looking here also with, uh, you know, looking at how the industry has evolved
01:02:32with AI, looking at the trends, um, what is the relevance of a portfolio in the industry, um, for
01:02:38people who are thinking about transitioning into the field? Portfolios are still really important.
01:02:43Um, I don't see them going away anytime soon. I still have a portfolio. I updated it a couple months
01:02:49ago with a new project that I built. Um, not for anybody. It was just me. I was just like,
01:02:54I had an idea and I built it and I never told anybody about it. I added it to my
01:02:58portfolio.
01:02:59Cause, um, even me, I need something to demonstrate my work. I have my YouTube channel. I have all these
01:03:05things, but I still work with, you know, uh, people and, um, portfolios are, um, I would say
01:03:15they're still extremely relevant in the fact that even with AI, you need to demonstrate your skills.
01:03:23They don't want to hire someone who doesn't know the skill that they're hiring for. It's as simple.
01:03:29It really isn't as simple as that. For example, like at a previous job, uh, I was on the hiring
01:03:37team.
01:03:38We would test for SQL or Power BI or, you know, whatever we were testing for. Um, and if they
01:03:44could
01:03:44demonstrate it in a project, it made them just in the conversation flowed so much easier in an
01:03:50interview. For example, with a Power BI, Power BI is a very simple tool on the surface. Um, very easy
01:03:57user interface, but using it in the real world is actually pretty difficult. That's why they're
01:04:01called BI developers, not a BI visualization specialists for the most part, because they're
01:04:07working with a lot of data pipelines and a lot of different things to create the data and make it
01:04:11good for the visualizations. And so when we would hire for these, these people, if they had a
01:04:16portfolio and we could just see what they've worked on and be like, okay, how did you do this? And
01:04:21they
01:04:21can walk us through that every single time someone did it. I could, I could tell right away whether
01:04:26they really knew what they were talking about or they didn't. If they didn't have a portfolio,
01:04:30it's harder to engage in that conversation and really understand what they know, what they don't
01:04:35know. Um, and so I have always felt and been a very big believer in portfolios and I don't see
01:04:42them going away anytime soon. I think they're still very relevant. And just to touch on what you said
01:04:48earlier, um, in data analytics, how, how much web development skills, uh, front end development
01:04:53skills, uh, come into play. Um, you mean like visualization? Yes. With data analytics. Very wide range.
01:05:02I literally have met people who are data analysts who literally only worked on visualizations. They
01:05:06were like, what, what I would call like a visualization specialist. They have the title
01:05:12of data analyst, but I've met people who do only visualization. I've also met people, including myself
01:05:19who did zero visualization for some of my jobs. Now, my first two jobs, I did data visualization in
01:05:25Tableau. And then when I became a data analyst at the fortune 500 company, I was doing, um,
01:05:32data collection. So we worked with, uh, you know, ETL pipelines and, you know, data migration and all
01:05:40these different things. I didn't work on the visualizations at all. That wasn't my job. We
01:05:44had BI developers. So then I would go to the BI developer. I'd say, Hey, you know, here's what we
01:05:49need for a client. Um, they would build it out. I would review their work before I hand it off
01:05:53to the
01:05:53client. I'd say, Oh, okay, let's actually change this. We need this data points instead of this.
01:05:57I didn't actually build it at all. So it varies widely. I think there is usually a happy balance
01:06:04somewhere in there for most data roles. Um, especially at small companies where you do
01:06:08everything, but at larger companies, um, there is a usually happy balance where it's like
01:06:1480% not visualization, maybe 20% visualization, or, uh, more often than not, I usually see it like
01:06:22it's 10% visualization or, you know, 15% visualization, but a lot of it is in the data
01:06:28using, uh, databases, working with clients, um, you know, different skills other than just
01:06:35visualization. It's usually a smaller percentage.
01:06:40Fantastic. Thanks for that. I'm just going to jump a very quickly. And I am mindful of the time here.
01:06:44Um, and we are pushing on, um, for anybody who's interested in maybe learning more of a tool,
01:06:48we have got a career foundry mentor who will be presenting another, uh, presentation in a couple
01:06:53of weeks, uh, Dr. Humera, who is a machine learning expert, um, based in Munich, and she'll be doing
01:06:59a deep dive into SQL. I've posted the link on a big market, but maybe someone from the team could
01:07:05just take that link and also post it on YouTube too. Alex, there was another great question that
01:07:09came in on YouTube from one of your audience, and it picked up on, um, something that you mentioned
01:07:14about clay, uh, cloud, uh, data analytics before Shane is asking, what are the skills required to
01:07:20become a cloud, a data analyst? Sure. I'm not surprised. That's a great question. I'm not
01:07:24surprised it came from my YouTube channel. They're very smart people over there. Um, what I will say
01:07:29is, is oftentimes most people don't start out in the cloud right away. Um, it, it, it just, I've,
01:07:38I've worked with a lot of people and most people who I've like either mentored or talked with,
01:07:41they're like, Oh yeah, we start off in SQL. And then, you know, then I moved to a company
01:07:46that used this, or then I moved to our company migrated to this. So most people don't start
01:07:50out in the cloud. Um, it just from my experience, but as more companies move to the cloud, like
01:07:57we were talking about earlier, I think it's going to be this, this year is for my YouTube
01:08:01channel. I'm going to be focusing a lot on the cloud. I'm going to have a whole series
01:08:05on AWS, a whole series on Azure. Um, because I, I, people need to start learning it for
01:08:11sure. Um, so what do you need to know? Uh, usually there's about three main components
01:08:17that I think are really practical to data analytics. One is something like Azure data
01:08:23factory. I'm going to speak specifically on Azure for this, uh, use case. Um, Azure would
01:08:28be for this example, Azure, um, does a million different things. Uh, it genuinely does. You
01:08:35go in there and start looking at Azure. It is overwhelming the amount of things that they
01:08:39have, but specific to data analytics. Um, Azure data factory is for like ETL pipelines and,
01:08:46and ingesting data. I worked with Azure data factory for like two years straight. Um, and
01:08:52you know, it's, it's a little bit more advanced. It's not something I would start out in that's
01:08:58I'll just, as a, you know, uh, a side note, I wouldn't start with Azure data factory, but
01:09:02it's something that one of the three things that I would learn. Um, the other thing is just
01:09:06working with their databases. So they have databases, they have data warehouses, they
01:09:11have data lakes. I would just start with the databases. If you know how to use the databases,
01:09:16usually you can pick up on how to use the warehouses and the data lakes are a little bit
01:09:20more complicated, um, with how you're actually working with the data and use it. Cause then
01:09:25you got to connect to something like Databricks and use some, maybe like something like PySpark.
01:09:28Um, but that's a whole different, that's a whole different conversation. Um, maybe we need to do
01:09:33another webinar on just like cloud platforms. I can do like a deep dive in like Azure and AWS,
01:09:37because that I could talk about that for a long time. Um, and then, and so, uh, I'm going to,
01:09:43I'm going to take it back. I'm going to say just those, I would just start with those two start
01:09:47with data where, uh, databases and data warehouses, just start with that, learn how to use it, learn how
01:09:53to put data into it from like an Excel file or connect to a data source. Just start out with
01:09:58that.
01:09:59Then learn about Azure data factory. Uh, that's where I would start. Now there's so many other
01:10:04aspects. Um, the Azure is a huge ecosystem, so it can be very overwhelming. I would just start with
01:10:10those two and then you can branch out into other things. Like I said, you know, there's data lakes
01:10:15and, and data bricks that's integrated in there. And there's just so many things. So don't get
01:10:20overwhelmed. It can be overwhelming. Same thing with AWS, AWS maybe even more so, but don't get
01:10:26overwhelmed. Just ease into it with a database. Like you should be somewhat familiar with. If you've
01:10:32used like my SQL or Microsoft SQL server, start off there and figure that out and then move on to
01:10:38the
01:10:38next stage. Awesome. I'm mindful of the time, but there's, there's some really great questions on
01:10:44YouTube. I think we should do a quick fire. I think we should do quick fire, Alex. We'll try it.
01:10:48Yeah. Yeah. We'll try this out. It's a new format. Okay. Shabin is asking if you were to choose only
01:10:56one data analytics tool for the next 10 years, what would that tool be? SQL. All right. Next
01:11:02question. Right. Uh, Cheyenne, sorry if I pronounce anyone's names wrong. Cheyenne is asking, what should
01:11:09I learn for front end data analysis, streamlit or dash? Uh, I like streamlit myself. I've used dash
01:11:16too. And it's also really good. I just have more experience with streamlit. Uh, I think it has a
01:11:20really good community around it. Fantastic. Gonzo is asking what are your thoughts on learning choral
01:11:26language before any other data language? I'm not super familiar with that. So I'm going to say pass.
01:11:36I think we should add the pass button too. That's good. Uh, zero. Do you think a Gemini ultra is
01:11:42needed to speed up learning a way or is it enough to use chat GPT 3.5? No, you can
01:11:48use, I would say
01:11:49even, even like chat GPT 3.5 is perfectly acceptable. Um, it does a very good job now. Uh, okay.
01:11:56This
01:11:57is supposed to be rapid fire, but you'll see improvements with chat GPT four and you'll, and it
01:12:01maybe if you're using the API, it's going to be a little bit different, but you don't need the latest
01:12:07and greatest with AI right now, maybe the few, maybe in like a year, you shouldn't be using 3.5,
01:12:12but 3.5 perfectly acceptable to using it's free. Awesome. And then Robert's asking, is learning
01:12:19Excel still worth, worth it given Microsoft copilot? 100%, 100%. In fact, I'm making a video right now
01:12:28on Microsoft, uh, copilot. Cause I, you know, started paying for it and using it. Um, it has a long
01:12:35way
01:12:35to go. And Excel is like the de facto tool in any company for businesses all around the world. It
01:12:42is,
01:12:42you have to know how to use it. So yeah, you'll see my video not too long on that. Maybe,
01:12:48maybe next
01:12:49week, maybe the week after. Um, but there's a lot of, there's a lot of issues with it right now.
01:12:55They,
01:12:55there, there was a reason why it was delayed for like six extra months or like eight months when
01:13:00they said that it would be coming out. There's a reason and it's, it's still there. Um, and so no,
01:13:06I'm not, don't, don't put your faith in GitHub, uh, uh, or, uh, Microsoft copilot right now.
01:13:13So Robert, this goes out to you. There's no way that you can avoid Microsoft Excel. You still got
01:13:18to learn it. You still got to learn it. It's going to be there. Um, Alex, I think we're going
01:13:22to end it
01:13:22there. That was some great questions. Thank you so much for the answers. Thank you so
01:13:25much for the presentation. As I said before, at the start, anyone watching on career foundry on
01:13:30any of our channels do go and check out, uh, Alex's YouTube channel, Alex, the analyst and help him
01:13:35get to 700,000, but also, um, he's got some great content up and coming. So do subscribe to that.
01:13:42If
01:13:42anyone's interested in data analytics, anyone's interested in career foundry, um, I'm just going
01:13:46to post on big market, a link to book a call with a program advisor. We've also put it in
01:13:50the
01:13:50description below. So do have a look there. If you've got any questions about jobs in your
01:13:54locality, our curriculum, the dual mentorship model guarantee that we have, um, do book a call
01:14:01with a program advisor and get your questions answered. We are currently offering a 20% off
01:14:06career foundry's data analytics program. Um, I've added a little sticky note on big market,
01:14:10but I think it's also on YouTube. So to claim that simply click there and, uh, it'll take you
01:14:14through to the discount, Alex, I'm sure we're going to see you again on the channel. This
01:14:18sounded fantastic. I think we should do some stuff about cloud platforms. Um, I think we
01:14:23should, we should go into some new territories, I think. Um, but thank you so much for presenting
01:14:28this evening and thank you to Alex's audience too, for joining us. And, um, we're going to
01:14:34be sending around the recording tomorrow by email, but you can also see it over on Alex's
01:14:39channel. If you get to the live section, you can see that. And also some previous webinars
01:14:44that we've done together, um, over the past 12 months. So, um, yeah, thank you everybody.
01:14:49And, um, it's time for dinner for me, but with some people, it's probably breakfast time or
01:14:53lunch time, uh, and Alex.
Comments