How to Start Your Career in Data Science
Scotland Data Science & Technology Meetup
In this meet up, Barry Smart shares his career journey: from software engineer, to architect, to IT Director, to CTO, and then going back to University to get a Masters in Artificial Intelligence and Applications.
Transcript
Bethany Rodgers-Rintoul: Up to about 45 and increasing, which is great. A very warm welcome to Scotland's Data Science and Technology Meetup today. We're going to be discussing how to get started in data science - how to set your first foot into that career. We've got two wonderful speakers, Barry and Delphine, who've come from quite different backgrounds and they've ended up in quite a similar kind of job position at the moment. They've come together to give an overview about how you don't need to follow one particular path to end up as a data scientist or to end up even in the data industry in general. There's so many different options that are out there.
There's so many different opportunities. So what we're aiming to do by having this event today is just give you a bit of insight into what we've had success with in the past, and to give you a bit of an overview into what's worked for us.
So to introduce myself, I'm Bethany Rodgers-Rintoul. I'm the project manager at MBN Academy. I've been working with MBN for about three years now, and I head up the DataLab MSc placement program. So I help to find industrial placements for students. I'm the main person within MBN who has links with academia, who helps just strengthen the relationships as we are - we're all aware that there is a gap between academia and industry, and it's really important that gap is plugged. And that's one of the things that we really aim to do.
I collaborate with the likes of CodeClan, Equate Scotland, and I'm quite close to as well the Open University, the DataLab - all these really fab organizations that are worth looking into. And they all exist to make Scotland, and the world, a better place, and to give people the knowledge and the impact that they need in terms of how to get involved in the data industry. And once you're involved, how to get even better and how to encourage more people to get involved as well.
So to get started today then, a few kind of ground rules before we kick off. Please ask questions. We're keen to get your input and make this event as interactive and as engaging as we possibly can. So if you've not used Zoom too much before, you'll see at the bottom of the page there's a Q&A and chat function. So please feel free to chat with us or individually, let us know as part of a group what your thoughts are on things that we're saying. But if you want to ask something that's more direct and something that you think would be useful for others to hear answered as well, then please use the Q&A function. I will be monitoring that as we go through the session, and I'll be asking questions when appropriate. If you don't feel comfortable asking a question on a public forum, then just private message myself, and I can certainly ask that on your behalf so you still get the opportunity to get an answer.
So yeah, please feel free to ask questions. The way that this event is kind of set up: Barry's kicking things off with a kind of 15-minute introduction. Then we'll move on to Delphine, and then round up with myself. And we've set aside around about 30 minutes, 40 minutes to have a panel discussion on the questions that hopefully you've had the opportunity to ask as we move along.
So without further delay then, let's kick things off. Our first speaker is the wonderful Barry Smart, who's going to get things started. So Barry's currently a consultant working with endjin, and I met Barry last year when I was coordinating the DataLab placement program. Barry graduated with a Master's in AI, but I'll let him do his own introduction and give you a bit of an insight into his data career. And he's always got a few very interesting hints and tips that he can share with you that I'm sure will be beneficial. So Barry, I'll pass on to you.
Barry Smart: Thank you, Bethany. Thanks for the invite to speak today to you and Michael. It's a pleasure. So all I'm going to do today is just tell my story. And so it's been actually quite interesting to have a think about where did my data journey start? And I've traced backwards from where I am today to think about it, and I think I've found the source.
I think it was about when I was in Primary 7 at school. My dad's an engineering professor - he's an academic. And so I was lucky enough at that point in time to get access to a computer. He brought this Commodore PET home, and you couldn't do much with it. You could program it in BASIC, and there were a few simple games you could play on it like Pac-Man. And you might recognize in the background there, the five-and-a-quarter-inch floppy disk drive - that was how you loaded programs into it. That's probably... the people on this call are probably too young to even have experienced five-and-a-quarter floppies, but they stored about 300 kilobytes worth of data. So probably couldn't save a small Word file on in today's money. But that was my early exposure to technology at that point. So I got, had an interest in technology.
The other thing I was interested in was playing the trombone. So these are my two best mates either side of me. So I was coming out of high school at this point. I was thinking - Drew and Jamie who you see there, they went off and did music degrees. One went to London to the Guildhall and the other one went to Leeds to study jazz. And I was at this kind of junction: do I keep doing music while I do something else?
Of course, I chose to do a degree in physics at the University of Strathclyde. I had a brilliant physics teacher at school and he inspired me to go off and do a degree in physics. So that's where I went off and did my undergrad degree. And as part of the degree, my final project was all about firing lasers at the skin and measuring the heat that's given off in the millisecond after that pulse - laser pulse hits your skin - and determining the thickness of the skin and all this kind of stuff. But as part of this project, I also developed a computer model to simulate what was going on in the skin under these conditions. So I was again doing a physics degree, but dabbling with the technology.
So once I graduated and it came time to think about what did I want to do, I had a choice to stay on maybe and do a PhD, or maybe to go into a physics-related job, which was kind of limited to defense-related stuff at the time. So I joined a company called Logica, who were at the time a global technology consultancy firm, and they had a graduate intake program. So I joined them in 1995 as a fresh-faced graduate. And I think the graduate intake - they took on about 30 graduates that year. They were a big company.
So I went through all the sort of training with them as a graduate, and I got involved in programming VAX PDP-11s. This is one such device I was working on. It looks really old - it was actually quite old tech at that time. They were using this technology to control water networks. So initially I worked for Glas Water and Yorkshire Water maintaining the software, but I then got sent off to install the software for Sydney Water in Australia and ended up being there for three and a half years working on that project.
And it was absolutely amazing working with the technology and applying it in that real-world scenario to help control the distribution of water to a whole city. And reflecting on that, this is all driven by data, isn't it? The data's flowing from the devices in the field into the sort of data center, and then decisions are being made - which valves to open and which pumps to switch on and all that kind of stuff. And those instructions are being sent back out to make all of that work. This is very much getting a taste of the power of data and what it can do.
So after that project, Logica opened an office in Edinburgh. So I decided to come back from Australia and work there, and that exposed me to Scottish Power. And eventually, having worked for Scottish Power as a Logica consultant, Scottish Power poached me and I joined their team, initially working in their trading department. So yes, Scottish Power had a commodities trading desk very much like you would see in the City of London - lots of screens and people shouting down the phone and buying and selling commodities. And they run a commodities trading desk 24 hours because electricity is traded round the clock. And that's what they did, but they also traded gas and coal and renewables and CO2 certificates and all kinds of things.
So they had a trading desk and I joined a team there as... quite fresh-faced. So I joined a design authority team and our role was to oversee all of the systems design for the trading department in Scottish Power. And you can see the maps on the wall behind - we were mapping processes, we were mapping systems, but all of these lines between boxes on these diagrams represented a flow of data through the organization. So in trading terms, you would have the front office - the guys on the desk buying and selling commodities - that would then need to flow through middle office so that the risk analysts could look at the data and understand the risks that they were exposed to: credit risk, market risk, et cetera. Then ultimately the transactions would also influence a back office so they could be settled ultimately in the market. So all of that was quite complex and it was just fascinating to be part of that team and to really act as the architect for that major multi-million pound program that they implemented.
So that went really well. And then I moved from that part of Scottish Power into the part of Scottish Power where power is generated. It was called Energy Wholesale. And I ended up actually being based here at Longannet Power Station. Now this is currently being dismantled, this power station. God, I'm getting old! But they're slowly dismantling this. The chimney's still there - it's the highest man-made structure in Scotland, but that's soon going to get demolished as well. But yeah, when I was working here, it was a coal-fired power station. So obviously not in fashion these days, but it was an amazing place to work.
And I worked as part of a team - using process safety team and process engineering team. And we, again, we were using data. Didn't quite realize this at the time, but we were using data to try and predict when equipment in the power station was going to fail. Because it's better to decide that a pump, for example, is going to fail and repair it proactively than to wait for it to blow up or fail catastrophically and do it reactively. So we were very much part of trying to shift into that more proactive maintenance regime within the power station. So that was a fascinating part of my career. But again, a lot of data flowing around to support all of that.
Then what happened next? Yeah. So then I left Scottish Power eventually and joined a company called Hymans Robertson. They've got... they're a smaller company, about 800 staff, and they were... they're a firm of consulting actuaries and investment consultants. But basically just think of them as a firm of professional mathematicians who help clients who are managing pensions or managing insurance funds to manage the risk around those things. So again, this was a fascinating organization to work in. In terms of those data, pensions data was flowing into the organization and then the actuaries would run models on top of that to calculate the risks and help advise clients on how to manage those risks.
And I joined them as IT director and eventually got promoted to CTO within Hymans. But part of my journey there was working alongside Microsoft because we recognized the opportunity to move onto the cloud to get access to the massive storage and compute resources that are available on the cloud. But of course, because we were working with personal data, we had to think really carefully about the security and all around all of that.
So this is me presenting at one of the Microsoft summits in London, talking about how we thought about the risks and eventually ultimately got board support to move some of Hymans' technology onto the cloud. And it was at this time as well that I met Howard and Matthew who founded endjin, and that's where I am working today. So this is where I first got to know those guys. And I guess a lot of the messaging today is all about building your network as you go through your career. So that's where that connection was first formed, and that was about seven years ago.
So then also when I was at Hymans, I met Gillian Docherty, who's the CEO at the DataLab. Got introduced to her and went along to some of the DataLab events. So I think three years on the chart, I went to the Data Summit which is the tail end of the DataFest that runs every year. And I was just blown away by what people were doing with data and just the whole atmosphere at the event was amazing. That's definitely one thing I would suggest is get along to these events, either virtually or hopefully soon we can go along to these things in person, and just... you'll learn a huge amount and it's also a great opportunity to network as well.
So there was one lady I sat next to who worked for the Highlands and Islands Council. Got chatting to her about what they were doing with data. You meet people when you're having your lunch, there's stalls, there's things like that you can go around and just generally meet people and get a sense of what's going on. So I started to get really interested in this growing focus on data and data science that was really being led in Scotland by the DataLab.
Then I got an opportunity within Hymans to run a data project. So you can see here, this is us - we ran a little hackathon. So one of the issues that Hymans had was that when it's bringing in pensions data from clients, it's often not in the best shape. There's data quality issues, or the way that the data's structured is not ideal for the way that the organization wants to work with it. And a lot of that processing was quite manual.
So what we ran here was a hackathon. We locked ourselves away in a room for five days, and we were joined there as you can see on the right by Jess from endjin. Howard from endjin was there as well. So we had an injection of expertise, and we had some practitioners from the organization. So there's Liam there - he was a developer actually. So stole a developer from one of the development teams. And we also had one of the pensions team experts in the room as well. And within the five days we showed potentially how you could take that process that took weeks and compress it down into a number of days, and how the new technology would allow simple but onerous tasks to be automated. And we were using some AI as well to detect patterns and do fuzzy matching in the data to clean it up as it went through this whole pipeline.
So that was a fascinating project, but the thing I really enjoyed about it was I got my hands dirty again. I was coding. I actually got involved in writing some code and I really loved that. And I just... it got me interested in... I'm missing that. I'm working at CTO level, I'm shaping the strategy, I'm leading the teams, I'm advising the executive on all of that stuff. But I just felt this was in my blood. I felt this was something I wanted to go back and do. I wanted to get my hands dirty again.
And I also realized that there's a lot here that I didn't understand. I could see the potential, but I still didn't feel like I was equipped really to lead on some of this stuff. Having chatted to my wife and my family about it, I made the big decision to take a career break. So I finished up at Hymans late 2019 and I started... I went full circle. I went back to the University of Strathclyde, but this time as a graduate to do an MSc in Artificial Intelligence and Applications. And this chap here, John Levine, was the course director - fantastic chap. And he immersed us in everything that is artificial intelligence. So we did a lot of data science, but a big focus for this course was about developing intelligent agents that could reason.
And one of the assignments that we had... and if you told me this is what I was going to be able to do at the start of the course, I wouldn't have believed you, but we wrote an AI to play the game of 2048. So I don't know if you've played this game, but you can do... it's popular in mobile phones as an app. And you can basically swipe on this grid - four-by-four grid - to move the tiles. So you can swipe left or right or up or down to move these tiles around. And you combine tiles that have the same value - they get combined. So you're constantly trying to compress tiles together. And the objective is to keep moving the tiles around - hundreds of moves, potentially - to get as high a value tile as you can on the board.
And it's quite addictive. I've maybe not explained it that well, but it's quite an addictive game and you can spend hours playing this thing to try and get this elusive high-value tile. So one of the assignments was to build an AI, and it made about three moves a second. So in between each move, it was thinking. It was trying to think, right, what's the best next move to take? And it was just amazing because I'd written that from first principles in Python and I'd written the code to do that. And I understood what was going on and I'd developed something that was thinking and actually doing quite well. I think it was generally better than a human would be who was a beginner at playing the game. So it would get quite far in the game before it would fill the board up with tiles, and at that point the game ends. So that was a fascinating project. So there was a strong sense of achievement in the project.
So that was great experience to come out with. So I was building, filling in the gaps at that point in terms of my knowledge about what is artificial intelligence? What is it all about? And what does it mean in practice to actually implement it and make it work? And this was a great example of how that knowledge was being filled in. I immersed myself in the course and got all of this value out of it.
Now, the thing I forgot to say was that I was sponsored by the DataLab to do the MSc, which was fantastic. And that was one of the reasons I was able to do it - was that they sponsored it. But the other thing about the DataLab sponsorship was that they also supported you right throughout the MSc and into the summer term where you were doing your sort of project. They encouraged you to do an industrial placement. So that's where I met Bethany because Bethany was responsible for finding us students a placement in the industry to do our project.
So I kind of used my network again at this point. So one of the chaps I'd worked with at Hymans had left Hymans and had ended up... in fact, two guys from Hymans that I knew at Hymans Robertson had started a new FinTech company in Glasgow called Nude. And so I got in touch with them and asked them, would they be interested in sponsoring an MSc summer internship student? And I guess because they knew me and that connection, that's what happened. So I joined them.
It wasn't a great time for them - they were very much in fundraising mode. They were... it was a big decision for them to bring me on board because they didn't want any distractions. There was only a handful of people in the company at that time. So they felt they were taking a bit of a risk, but I'm pleased to say that it worked out and I got access to some brilliant data. So open banking data, and I was able to develop some algorithms to start spotting regular spending patterns in people's bank accounts. Because a big part of Nude's value proposition was to help young people identify, by looking at their open banking data, identify areas where they could potentially be saving money. And that was the kind of start of that kind of data science-driven insights for their customers to help them get into their first home that bit quicker by identifying those opportunities.
So that was a fascinating summer project. And it was just absolutely fascinating to be part of a startup at that point in its journey. And while I was there, they managed to secure funding and they started to grow the team and they're absolutely flying. They're worth checking out - brilliant organization, great idea that they're taking to market. And it's really some fresh thinking that they're bringing to the banking industry as well.
So that was great. And then again, just working my network when it came to finishing up my MSc and looking for opportunities there, I got in touch with Howard from endjin, who's one of the founders of endjin. And it wasn't really to ask for a job at endjin. I was just keen to get his ideas about a few kind of opportunities that I had. And then the conversation turned to "why don't you come and work for us?" So having been a client of endjin, I ended up becoming a member of the team and that's where I'm working today. So again, it was that kind of just using your... the network that you've built throughout your career. Get in touch with these people, tell them what you're up to, and you never know what could happen. What could come out of this?
So what does endjin do? Well, it's a small business and technology consultancy that specializes and has a great deal of expertise in the Microsoft Azure cloud platform, big data, artificial intelligence, and complex software engineering challenges. And that's effectively what we do - is we work with clients and help them to realize the full value of data and technology, in particular leveraging cloud technology. And we work really closely with our teams. We bring expertise and know-how and some intellectual property that really helps them accelerate their journey and generate a lot of value from it.
So I've been working with endjin for just over three months now. We've done a bunch of projects already in that time. One with a firm of financial advisors, and that was... we helped them to move up the data maturity curve quite significantly. We're currently working with one of the world's biggest sporting brands and their focus is on knowledge rather than data, which is fascinating. And we're also working with a big marketing firm where the data volumes are just eye-watering amounts of data streaming in from social media channels. And we're working with them to generate value from all of that data.
So there's some really interesting projects and it's a nice broad range of skills that... it's all my 25 years of experience working in the tech industry, but as well, I'm getting to apply a lot of the new skills that I learned on my MSc as well. I'm finding it really challenging, still learning a lot, and enjoyable just to see the impact that we're making on the clients that we work with.
So there you go. That's my journey. Hopefully, Bethany, that didn't run over time too much and hopefully take some questions later on. Thank you.
Bethany Rodgers-Rintoul: That was brilliant, Barry, thank you very much. I've just written down a few kind of key points that you mentioned that I think are so relevant for people that are trying to get involved in the data industry. I remember speaking to you probably this time last year, and you saying "I don't have that much of a network." I think that's often something that people massively underestimate. Yeah. Especially if you come from a different background or a different career, you will know people that can help you take the first step. And definitely take advantage of that. So that's brilliant. Thanks very much, Barry.
Barry Smart: No problem.
Bethany Rodgers-Rintoul: And a few great questions came through, so we'll get to them a little bit later. So I'm now going to pass on to Delphine. How are you getting on with sharing your screen? Are you able to do... do you want me to pop it up for you?
Delphine Rabiller: I think, yeah. If you can pop it up for me, that'd be great. It's only slides, it should be quite quick.
Bethany Rodgers-Rintoul: Okay, good. No worries at all. Okay. Can we see... that's okay. Yeah. Perfect. Perfect. So just tell me when to move on, Delphine, and I'll move things on.
Delphine Rabiller: Like to be about me slightly. That'd be great.
Bethany Rodgers-Rintoul: Perfect. So thank you very much, Delphine. I'll let you take it away then.
Delphine Rabiller: Thank you. Thank you very much for inviting me to speak. Yeah. I'd like to share my journey in the hope that it's going to be helpful for somebody else. So if we go to the next...
I was going to chat a bit about where I come from and how I got into data science. My journey's not that straightforward. I didn't know how to program until quite recently. I've always loved science. Since I was really young, always loved maths, always loved physics, always loved chemistry. It was quite clear for me to go into an engineering degree. So I studied chemical engineering in France and got loads of hands-on experience in that field. But my English was quite appalling at the time, and to do science in France, you need to be able to speak English. So I decided to go and do a PhD in organic chemistry, moved to Birmingham, and then followed my supervisor to Glasgow.
And following my PhD, I had what I would say was quite a successful career in biotechnology. So for 11 years I worked first as a scientist and then a senior scientist. And then I got promoted to supervisor role. And I ended up in my last role where I was leading client projects. So customers would call us and say "we would like to investigate this problem." And we would take the project as our own and define it and work on it, try to find solution, how to answer the questions, analyze the data, and then provide the report. And I really loved that job. And I got to a point where I was managing the lab and I was following the flow where I was progressing in my role, the way I was supposed to progress.
And I remember, I vividly remember that moment where it was appraisal time. And that question came up: "Where do you see yourself in the company in the next two years?" And I was like, "Actually, I don't think I want to be here anymore."
And it was a key directional point for me where I was thinking, actually, what is it I like about the job that I do at the moment? And what is it that I like less about it? And I love... I've always loved problem solving and that's why I always loved science. What I like in my current role is I get a problem or high-level questions from a customer. I translate it into sub-questions, do the experimentations where I need, find the data that I need to be able to answer that question, analyze the results, and translate it into something that the customer can understand. I absolutely love this. Turns out that job's got a name and it's data science.
Now this is all I want to do. I don't really... doesn't really matter what the rest of my job is. This is what I want to do. If we go to next slide, Bethany. And that's where I decided to ditch the test tube and try to see how I can get into data science. I left my job and then I had a lot of time to think how to get into data science. Because some would say, and I say this to a lot of friends that come to me to ask how they can get into data science as well, I was already a data scientist. The two things I was missing was I couldn't code - never used the computer to do anything ever before. And there's only so much you can do with Excel. And my network in data was very small. All my network was in science environment. So I needed to grow those two areas for me.
I talked to a lot of people who'd done career change. All of them said, "Yeah, it's going to be hard. But it's totally worth it because you're going to be so happy once you do what you want to do." I shadowed a few people that I knew that were working in data science role, just to see if actually it was what I wanted to do. I also went to speak to Equate Scotland who does help people - women, especially - who want to change career. And then that's when I realized that all the process, all the logic that a data scientist must have, I already had this. What I needed to work on was the coding aspect of this.
So I started to work on my own and followed some Python tutorials, started working on a portfolio just to have something to demonstrate. And I also got in touch with CodeClan. At the time, CodeClan only had their software development course, but they said they were doing pilot courses for different fields. So I was like, "Well, harm to ask. I'm just going to contact them, check if they did anything in data science." And I got a reply to say, "Funny, you're asking. Turns out that in a month's time, we're just about to launch our data science course." So I went to... and we basically, we were filling all the gaps that I had in my skills. They were going to teach me how to code using R, teaching me a bit of Python, teaching me a bit of SQL. But most importantly, provide me with a network of companies that would be looking to recruit people at the end and where to find that network that I was missing. So I decided to sign up for this.
Again, it was hard - three months of very intense learning, but it is really all worth it. We go to next slide, please. Bethany. Thank you.
So one of the things that I learned that was really key to, especially to where I am now, is network is everything. In Scotland there's a brilliant tech community that meets every month. And so those are a few meetups where I've joined: Women Who Code, PyData. I'm also sometimes volunteering for a charity called DataKind who provides data help for social organizations. And the reason this is so important is that it gives you the opportunity to talk about what you do, to see what other people are doing, keep up to date with what everybody else in the community's doing, latest skills, latest technologies. If you're looking for a job, who's recruiting - that's really important as well. And it opens so, so many doors.
So the reason I'm saying this is because networking is how I got my current job. If we go to the next slide. So now I have a job I absolutely adore. I work for a company that is called Massey. And when I first started at Massey, I had a meeting with my manager and we went outside and had a walk in the park. She was saying, "Yeah, when we posted the ad for your job, we had 450 applicants." And I couldn't believe it. It was insane. I was like, "Why me?" And what made the difference is because, so during CodeClan, we had done some user testing with Massey that had come to talk to us. We had spoken to people who were recruiting for my role. Martina's also leading the PyData meetups. So when I gave a talk at PyData, I had the opportunity to meet with her. She knew how I worked and just networking as well, opened so many doors to you if you take the chance to know people that are working in the same field as you.
So my current role as a data scientist in fashion, I've never thought I would have worked in fashion, but I find it's basically about helping... around everybody in the company. Everyone works with data at Massey. So it's about helping modeling scenarios for our finance team, it's about helping our buying team to make better choices for their collections. What sort of patterns, what sort of fits, what sort of colors do users buy? So it goes from user analytics to broader productions. I learn stuff every day. I learn new technologies every day. And for a job that I've done remotely since I started, it's been a delightful surprise to be where I am at the moment. And yeah, that's me.
Bethany Rodgers-Rintoul: That was great, Delphine, as always. Thank you very much for that. 450 applications for one data scientist job.
Delphine Rabiller: It was advertised as remote. So we had applications from just about everywhere.
Bethany Rodgers-Rintoul: How important do you think your established network or the groundwork that you'd put in, how important do you think that was in helping you land that position and beat the other 449 people?
Delphine Rabiller: A hundred percent. It was a hundred percent based on that. One other thing they told me is "we saw your CV, recognized your name. We know you, we'd seen you." And my coding skills are not the best, and I'm still progressing on that front, but yeah, the fact people know you, it makes a world of difference.
Bethany Rodgers-Rintoul: Absolutely. A hundred percent. And I loved your comment as well about how you've been in this job for quite a while now. And I'm sure you're the same as well, Barry, in terms of you can be doing the same thing, but you're always learning new ways of adapting and new technologies. And the world of data science and data is changing at such a fast rate. You're never really going to be stuck in a rut. There's so many opportunities to allow yourself to develop.
Barry Smart: Yeah. And I think that's key that you've got to be prepared. It's lifelong learning that you're setting yourself up for because it is changing so quickly all the time. So yeah. You're always learning new things, which is great.
Bethany Rodgers-Rintoul: Exactly. Okay. And that kind of leads me on quite nicely to my presentation as well. I'm going to try and share my screen and I can see we've got some brilliant questions coming through as well. So we'll get right to that. Are we okay? Can we see at the moment? Yes. Okay, perfect.
So to be honest, I'd set up my part of the presentation to cover the power of social media and networking. And I think just the things that Barry and Delphine have said, I think it really emphasizes that even further - the importance of it and how it can really make a massive difference to establishing yourself as an unknown quantity within the data community, regardless of whether you want to be a data scientist, a data analyst, a data engineer, regardless of your job title.
I think the data community within Scotland is so small, but as I'm known, Delphine and Barry would back me up on this, people are very genuinely helpful. When you reach out to them, when you ask for assistance, help, advice, when you need to grab somebody for 20 minutes, half an hour to find out how they got established in their career, nine out of 10 times, they say yes, and that's really what I'd encourage everyone to do.
So a brief summary of my session just now. I'm going to cover the importance of data literacy, where to find opportunities and what employers look for. Again, everyone in the audience, please keep the questions coming in. We've got about 18 that have come in so far, which are fab. I've also put my email address up at the top as well. So if you do have any questions after the session, drop me a note, connect with me on LinkedIn. Sure, Barry and Delphine are more than happy to do so as well. So keep in touch with us, just as this presentation will grow your own network by attending events like these and connecting with us afterwards.
So to get things started, what is data literacy? Data literacy is the ability to read, understand, create and communicate data as coherent pieces of information. It's about understanding how to ask the right questions, what the right questions will be. And I loved what you said earlier on, Delphine, in terms of you were always quite naturally curious about problem solving, about knowing the questions to ask, to break that down into subsets of different questions. You were a data scientist all along, you just didn't know it. So data literacy is so important. It's about being able to build your own knowledge, to make decisions and to communicate meaning.
A data scientist is... Delphine and Barry have correctly kind of described... there's so many different definitions, but through my experience of working in the data science and technology, and just over the last couple of years, as someone who can take complex large pieces of information and break it down so people like myself can understand it. It's about being able to translate these complex requirements and presenting it to stakeholders in the business. It's really important that's a skill within everyone. And it's something that can be naturally acquired over time as well.
This is just a piece of information that I got from DataIQ, an organization that are worth looking into. They list the kind of hundred most influential people within data. And they talk about some really fab things that data organizations are doing to increase innovation and make the UK a more advanced place. And I thought these were worth mentioning. So only 32% of companies reported being able to release tangible and measurable value from data. This is partially due to individuals not truly understanding data and its potential. And that's taking... it's not specifically data companies, that's companies in general throughout the UK. I think it was a thousand or something like that, that they... 32% of companies is a shockingly small number.
The Human Impact of Data Literacy report, a recent study commissioned, reviewed only 21% of the global working population reported that they're fully confident in their data literacy skills with one quarter prepared to use data when entering their current role. The vast majority of employees, 87%, recognized that data in the workplace as an asset, yet many firms are struggling with how this is the kind of best linked with the people, the process, the technology elements necessary to truly establish a data-informed culture. 87% of employees recognize data is bringing massive value to a company, yet only 32% of them are able to draw tangible or measurable value from the data that they've got. There's obviously a bit of a disconnect there, and I think this leads to a bit of a continued data skills shortage with organizations wanting to embrace the data trend, to increase their insight, to be able to make better business decisions, not necessarily knowing what to do. And that's why it's so fabulous for the courses that we've just discussed. And as I say, these are just a handful that can offer individuals that want to enter the data sphere, the opportunity to learn how to do this. There's organizations that offer upskilling for organizations, upskilling for individuals that can really help meet this requirement and this gap that we're seeing.
In terms of where to find opportunities, I can break this down into distinct areas. Having a good quality CV is so... having the appropriate information that's included in it. People that know me from my work with CodeClan or the DataLab will know that one thing I harp on about is a CV doesn't need to be two pages. As long as the information that's on it is concise, it gives people a really good understanding of what you're doing. But more importantly, what you want to do moving forward, a CV should be quite future-facing and it should give the reader an indication of what your intentions are moving forward. Don't assume that people know what you're doing. If you put down that you're doing a Master's or a PhD or you're retraining yourself, you need to explain it to them because they might not necessarily know what that particular course is. Especially with universities and training organizations, introducing wonderful new courses. And again, I'm going through this quite briefly, because I'm conscious of time.
But networking, as we've seen from Barry and Delphine, networking is so important and Scotland has such a vibrant and friendly community of different events and different socials that you can take part in. The one that we are joining today is part of Scotland's Data Science and Technology Meetup. This is one of many. Look at what works best for you. As Delphine mentioned earlier on, Equate Scotland, have an amazing network that can really help individuals get the kind of first step on the career ladder that they might need. Look at the organizations that are out there and join the ones that are best for you, but do it consistently. Just because you've attended one particular meetup doesn't mean that people will necessarily know who you are or remember it. You need to really make an effort to be known when you attend these networking events. Tag individuals on LinkedIn and things like that afterwards, ask them questions during it, try to be quite interactive and you make a bit of an impression. And that's always a good thing. If you're remembered and people see you as being a bit of an influencer or someone who really came to get involved in the data industry, you don't know who's watching.
LinkedIn expansion is so important. My kind of best hint and tip here is if you're looking to get established in the data science and technology industry, and you don't necessarily know where to start, I would write down a list of organizations, think of perhaps 50 or 60 organizations that are operating in a similar space of where you could potentially see yourself going into in a couple of years or a couple months. Look up the people that are working in that organization, both in terms of technologists, data scientists, data analysts, data managers, whoever potentially may be. Connect with them on LinkedIn and send a personalized request. There's nothing worse than receiving a standard blank request from people because you don't know why they want to connect with you. And it doesn't really help spark a conversation. Whereas if you send a personalized request, it gives the person a bit of an indication in terms of who is this person, who is this guy or girl, but also if they accept that, it pops up in the chat box and they can't really ignore that, it's something that they need to respond back to.
Social media so important as well. I would always recommend the two main ones of Twitter and LinkedIn, but again, just because you're on it doesn't mean you're automatically going to get flooded with opportunities. You need to ensure that you're being active on it. You're putting out relevant posts. You're putting out things about thought leadership. You're connecting, you're messaging, you're growing your own network naturally. It's a slow burner. You're not necessarily going to get an opportunity overnight, but as long as you do so consistently, it would really work to your advantage.
This is just something that we made a few months ago that I think is quite relevant about why should we network. If you think about it, and just to the point that Delphine said earlier on, 450 applications for one data scientist role in Edinburgh, there's not going to be much that kind of differentiates a lot of folk. Whereas if you can put down that you're an active contributor to Scotland's Data Science and Technology Meetup, you've attended DataFest three years in a row, but more so if you know the people through doing that, it's going to give you an even bigger advantage and really going to make you stand out from the crowd a lot more.
Again, for individuals that are thinking about entering this industry without any previous experience, by attending these events and these networking socials, it allows you the opportunity to build up your own confidence. You're not going to be an expert in the industry and people don't expect you to be. If you're attending events that are talking about the latest development in cloud computing or what the perfect analyst looks like and things like that, it allows you the opportunity to build up your own knowledge and network with people who can tell you a little bit more. People like talking about what they're good at, and it's a great opportunity for you to connect with them.
And most importantly from a recruitment perspective, the majority of positions are unadvertised. Around about 70% of all jobs never hit a job board. If you think about it, recruitment can be a very lengthy and a very expensive process. And I'm sure Barry and Delphine, they've been in situations like this in the past, where if you're looking to hire within your team, the first thing you do is ask your team. Does anyone know anyone? Is there anyone in the organization that we could potentially upskill that you think could be a good match for this? You attend networking events to see if you can find people. If you're known, if Delphine has been at networking events for the last year, and you can see that she's really passionate, she's going to get a tap on the shoulder. It saves a lot of time, effort and money. And just as we were all talking about earlier on, communication is massively important. And by attending networking events, it allows you to build your knowledge through personal interactions and develop those crucial soft interpersonal skills, which as we all are aware of are so important in this industry.
Now, the two main ones, as I said earlier on are LinkedIn and Twitter. I think Twitter is brilliant because it gives you the opportunity to reach out to folk that you wouldn't necessarily have the option to do so elsewhere. If you think about it, when the world was a normal place before coronavirus, if the chief data officer at Barclays or HSBC was given a presentation, there'd be a huge queue outside waiting to meet them. And you wouldn't necessarily have the time or wouldn't necessarily remember it because there's so many interactions and results in that day. Whereas if you connect with someone, if you follow them on Twitter, if you reach out to them on Twitter, basically the equivalent is sending them a letter. It's very much in their face, they can see it a lot more obviously. And they're going to remember your interactions if you're doing them often and often you're tweeting and talking about relevant things, the same with LinkedIn as well.
Again, conscious of time. So I'm just kind of whizzing through this part here. And it might be useful actually for Barry and Delphine to have a bit of a conversation around this in terms of what employers look for. Because I know that a lot of the questions that are popped up around this from my perspective and I can go over my main points just now. And I'll invite Barry and Delphine to jump on again with their thoughts on this as well.
The biggest mistake that I find in CVs and speaking to people is that there are some individuals... they know exactly what they've been doing. If they've undertaken a Master's in AI or there's a brilliant Master's at Strathclyde University called Machine Learning and Deep Learning. And that's a new course. Having a title like Machine Learning and Deep Learning, it doesn't tell me much. Same with Artificial Intelligence. It's quite a niche area and you don't know exactly what's going to be encompassed within those few words. It's quite difficult. So I would say on your CV have around about a third of a page detailing exactly what your retraining course is. If you're undertaking one, it's so important to give individuals a bit of an insight, especially when the job market is so competitive. People aren't necessarily going to have the time or the motivation to look up exactly what it is that this means. So I'd really encourage doing that.
Just as Delphine was saying earlier on, communication, having structured thinking and natural curiosity. Those are so important within a data sciences position. And it's something, again, going back to the networking side, if you're networking and using social media to the best of your advantage, you're going to naturally develop these skills because they're difficult and they don't come naturally to everyone, but you need to make an effort to try and develop them.
A data scientist is someone who's flexible, always learning. And in my experience in recruitment, I think if you've got a good underpinning within SQL to draw insight from databases, then that's a very good start. And having experience in analytical tools, such as R, Python and Tableau, having a bit of an underpinning in that is always going to be an advantage. Again, a common mistake that I see on CVs is people listing R, Python and Tableau as a skill. And those are programming languages, the skill is programming, the subset is having experience in that particular area, because if you've got skills in programming and it's great that you've got experience in R, an organization might be working in Python or might be working in another language. But if you can learn one, it's quite often that you can learn another.
Technical skills. Again, they're so important to have. And it's brilliant if you can display them. What I love about individuals that are entering the data science or technology market, if I'm a recruitment person, I'm a project manager person. If I look at having a new opportunity, I can't pull up a GitHub page and say, "look at what I've been doing. Look at all this fabulous stuff." Whereas people working in this market can, if you can start a GitHub page and put up examples of exactly what you've been doing, it gives people a bit of an insight. If you're retraining yourself and going through different training courses, then that's absolutely brilliant. Record them. Get them somewhere on a repository somewhere so that you can visually show examples of your work and the level of your competency.
And this is a quote that I found somewhere, just saying that technology is the enabler. Being able to translate and communicate the results is very important. So definitely I think that just underpins everything that we've been saying earlier on. Delphine and Barry, I don't know if you have anything to add to this as well?
Barry Smart: Go for it, Delphine.
Delphine Rabiller: I actually think this is a brilliant summary. It just varies... it's like a to-do list as a data scientist for how to get into data science. That's it.
Bethany Rodgers-Rintoul: Good. Good. No, that's perfect. What's your thoughts, Barry?
Barry Smart: I think another thing to do is if you get a chance to talk to a potential employer, do your research on them and show a genuine interest in their business and what they're trying to do with data. It could potentially convince you, you don't want to work for them in some cases, because they might be a bit vague because you want to go into an organization that's really wanting to get lots of value out of you as a person. And it can be sometimes demoralizing for some data scientists, because they can go into an environment where they're the first data scientist and the business hasn't really figured out how to get value from them. So protect yourself and it'll help you get through the interview process and show that kind of questioning mind you want, understand their business and how they're going to leverage data.
Cultural fit's another big thing. So they'll be looking for someone generally that's got a lot of energy and is interested. And I think that actually counts for a lot more than just the skills listed on your CV. Is this someone that's got all the skills that we're looking for in terms of programming languages, it might be different, but are they going to be able to apply themselves and adapt and fit in with the rest of the team? So try and bring out, you'll have wider experience of all of that in other roles that you might have had. So try and bring all that to bear as well. It's not just the data science skills that count, it's the wider life experience that you've had, that can really help you get over the line with these kind of jobs.
Bethany Rodgers-Rintoul: Yeah, absolutely. No, that's great. We've got a good handful of questions that have come in, so we're going to try and just work our way through some of them just now, if that's okay. So I'm just going to start at the bottom and work my way up. Okay. So what networking or meetup events can you recommend?
Maria, I would personally join meetup groups that Delphine was talking about earlier on and just have a browse. There's so many that are out there and pick the one that's best suited for you. From my perspective, I think that Scotland's Data Science and Technology Meetup is great. It's running in collaboration between endjin Solutions, the organization that I work for, and the DataLab. There's events around about twice a month. More than that on some occasions. And they're in general quite well attended. They are all recorded and put on the endjin Solutions YouTube page afterwards as well. So that for individuals that can't attend, they still have the opportunity to do that. What about yourself? Do you, Delphine and Barry? Have you found anything be particularly useful or beneficial for you?
Barry Smart: I would say the DataLab, DataFest events - get along to those. There's a whole range of events there. And it's on again this year, it's canceled last year, unfortunately, but it's on again this year. So look that up and get along to any of the events there that catch your interest. It's a fantastic opportunity to network.
Bethany Rodgers-Rintoul: Definitely. The DataLab are actually doing free tickets at the moment as well. So it might be worth people having a look at that. So it's called DataFest. And it's a kind of festival of data. If you didn't get that. It runs for a good period of time. I think that you said that you'd been at it for three years on the trot, Barry.
Barry Smart: Yep. Yeah. And there's fringe events that run - different organizations run fringe events that might really catch your attention as well. So look at the full program and go along to the things that look good.
Bethany Rodgers-Rintoul: Yeah, definitely. There's a question from Andrew talking about both of the speakers had years of experience in various sciences and engineering that most graduates wouldn't have, which experience or what experiences or skills do you think would be most useful for going into data science? What are the technical skills that graduates should focus on when it comes to looking for junior data scientist roles?
So from my perspective, Andrew, I think the technical skills are obviously very important and being able to show them is crucial when it comes to meeting up with organizations or trying to get a foot in the door. From my perspective, I think as long as you have a solid underpinning in a programming language, whether it be Python or R, that's really important. Being able to display your SQL skills in order to extract data from whatever it's stored at the moment is obviously very important as well. And also playing on your communication skills and talking about, if you're a graduate at the moment, talk about what you've graduated in. It might seem absolutely stupid of me saying this, but genuinely you'd be surprised at the amount of people that don't put information underneath their graduate degree. What about yourself, Delphine and Barry? What do you think a graduate should focus on?
Delphine Rabiller: I would say building a portfolio is quite a good way as well to differentiate yourself. It can be any size of project you want to do. Just a small data project where you can share on GitHub and show your logic, how you work from the high level question to actual results. Show your way of thinking and show whether it's a kind of screenshot of a poster or a dashboard, or how you would present this to somebody that is interested in just show your way of thinking is really important.
Barry Smart: Yeah, I would agree with that. Another good platform to look at is Kaggle. So the nice thing about Kaggle is they give you data sets that you can play with and you can actually see how other people have approached it. So if you're getting started with some of these skills, you could potentially take someone else's project and try and improve it and talk through how you've taken it and understood it and tweaked it a bit to help make it even better. And that's a nice place to showcase some of your work as well.
Bethany Rodgers-Rintoul: Absolutely. And I noticed that Ryan has just commented in the chat box. A question that I think's really relevant: "What's defined for opportunities. Does the work experience in other domains other than data science add value to the application, or should we only highlight the skills required for data science?"
Ryan, ultimately every domain is dependent on data, so you've answered your own question there. Absolutely. I think you need to be conscious about how you're presenting your CV and how you're presenting yourself. When it comes to talking about your previous industry experience. I spoke to someone yesterday, who's graduating with a PhD in environmental sciences. They're going on to do a data science type role. And on their CV, it was talking about tutoring and it was simply one sentence along the lines of "I tutored people." And I was like, "No, you need to explain that more. You've developed your communication skills. You've been able to work under difficult circumstances. You've been able to develop X, Y, and Z." So being able to draw out the experience that you think is relevant is so crucial. And ultimately people need to understand the background that you've come from because it's really going to help them, find that perfect position for you and optimize your previous industry experience and how that can be adapted to yourself. I suppose, Barry and Delphine are really the best people to answer that question in terms of coming from other industries. What's your thoughts?
Barry Smart: No, absolutely. Are you a team player? I think the other thing to note about data science is it doesn't start with the data. It starts with the business problem. So again, coming back to that, inquisitive mind, really trying to solve, understand the problem that you're trying to solve rather than, and your ability to do that in different situations. And the solution might not have been a data science solution. It might have been something else that you've done previously in your career, but that problem solving mindset, I think is probably the most important thing to get across.
Bethany Rodgers-Rintoul: Yeah, absolutely. There's a fab question that's actually come up a few times in the chat here. John, who did, has been the kind of first one to ask it that I think is so relevant and we get this all the time. "I'm in the position of trying to learn data science at the same time as being the only person in my organization, working within this route. I don't have anyone in work to learn from, or to talk ideas to. Meetups like this are great, but they're often high level. I'd be looking for a more practical hands-on meetup or perhaps informal mentors. Do you have any recommendations, John?"
From my perspective, the Scottish data science and technology area is so friendly. If you ask, I've got a kind of informal mentor that I collaborate with, and I know there are senior guys and girls within the industry that have exactly the same. If you are networking and meeting people, then it gives you a really good opportunity to make bonds with them, make friendships with them and ask them if they can give you some advice or some mentorship experience. Surely Barry, and Delphine you came across that within your careers.
Barry Smart: Yep, absolutely. And John, get in touch on LinkedIn - happy to have a virtual coffee with you at some point, so happy to help you out there. Absolutely. And there'll be lots of other people in a similar position willing to spend the time and help people. Keep chasing up those connections.
Bethany Rodgers-Rintoul: A hundred percent.
Delphine Rabiller: There's also something called Code Bar. And I used them when I had my first role in data after CodeClan because I was also the only data scientist and sometimes I was really lost and didn't know what to do. Code Bar is really good for that, with that small tutoring, where you come with your own questions and you've got somebody that spends the evening with you trying to go around it and explain you how to do things and it can be as high or low level as you want.
Bethany Rodgers-Rintoul: Definitely a hundred percent. There's a few questions again that are just talking about networking groups, but I think we've talked in detail about the different types of opportunities that have worked for us. Go on the meetup group or even just search on groups on LinkedIn. One thing that I love about LinkedIn groups is you can join up to 60 and when you're connected with people in that group, you're automatically kind of connected with them in terms of you've got the same exposure as if you were first degree connected, so you can message people without being connected to them. If that makes sense. It allows you the opportunity to pop in people's inbox without being first degree connected to them. And it allows you the opportunity as well to expand your reach. There's thousands of people on these groups on LinkedIn and you don't necessarily know, who's seen your post and who's thinking that you're doing some really good and cool stuff. So it's a really good opportunity for you as well.
So Steven has asked, "Both guests have moved from senior positions to perhaps less senior ones. What would they say were the best and worst aspects of changing direction in this way?" That is a fab question. Steven, Barry, what's your thoughts on that one?
Barry Smart: It's been liberating to... yeah, you're right. I've taken that step into sort of from a senior role down but it's much more hands on and it just appeals to my problem, solving instincts and to get closer to the problem and be part of properly solving it. So I'm really enjoying it. And it was just, to me, it was all about being in a job that I enjoy. So yeah, it's been a great move for me. Absolutely. Definitely.
Bethany Rodgers-Rintoul: Delphine. What about yourself?
Delphine Rabiller: For me I would say it is... same as Barry, liberating is the keyword. I would go from doing the job of five people to doing the job of one. I have time for myself, which is something that is new. I can spend the time... I'm passionate about what I do, which means that it's not a chore anymore to keep up my skills and to do courses and to learn new programming language and new things. The question I get the most is "what about your salary?" Yes, my salary's a lot lower than it used to be. But it's little in comparison to what I gain.
Bethany Rodgers-Rintoul: Absolutely. Absolutely. It's I suppose it's about being happy as a person at the end of the day as well. And seeing that there's going to be a clear career path for you moving forward. If you work with the right organization, as long as they can set up something that assures you, that there is going to be progression. And at the end of the day, you could be on a job, that's paying you an absolute fortune, but you might hate it. It's about doing something that you love, which is great.
An interesting question from Laura, "how hard is it to get into data science without a degree?" And it's a question that we get all the time at endjin as well. From my perspective, if you're wanting to work with a larger organization - JP Morgan or a Barclays or an organization that normally requires a Master's or something along those lines, it is going to be more difficult for you. There's no denying that, but I think if you are able to, as Delphine was talking about earlier on, if you're able to produce a portfolio of work, if you are taking part in Kaggle competitions and data hackathons, if you have portfolios available on GitHub and repositories like that, it really massively helps you. And I know I'm going on like about this, like a broken record, but with smaller organizations that are a bit more flexible, a bit more adaptable, if you can show concrete examples of your work without going through university or a training program, it's brilliant. And quite often I'm sure, definitely Delphine and Barry would back me up on this. There's individuals that have never went to university for a day in their life that are absolutely fabulous. You don't need to go to university to get into this field. In some instances, of course it will help, but there's obviously going to be ample opportunities for individuals that can't afford to go back to university or to take part in a retraining course, there are opportunities out there, but I do think you need to be better prepared in terms of having a portfolio of work, having a brilliant network and knowing exactly what's happening with the latest trend in technology. What about Delphine? What's your thoughts on that?
Delphine Rabiller: Very much the same. And the reason I did CodeClan is because it all depends on how you learn the best. If the way you learn the best is by doing then by all means, just build up your portfolio and show how amazing you are and what you can do. I need more structure. I needed to have a course where I had my homework. And yes, it was intense, but that's how I learned. As long as you got, whether it's a degree or portfolio, if you can show what you can do, you don't need a degree for it.
Bethany Rodgers-Rintoul: Absolutely. Absolutely. And I know Barry, obviously you've been in a position in the past where you've been in a hiring capacity. What would impress you, or what would you like to see from an individual who didn't necessarily have a degree or a formal qualification? What would be something that made them stand out?
Barry Smart: Yeah, it's just the passion and drive for the role. And the fact they've may not have done a university degree to pick up the skills, but they've been putting their own time in to build up the skills outside of that. That kind of thing I think counts for a lot.
Bethany Rodgers-Rintoul: Absolutely. And that links into Garys question about how many self-taught data scientists have you come across? Are self taught applicants at a disadvantage in regards to how they're looked at during recruitment process, or is it purely skills based? What's your thoughts on that, Barry?
Barry Smart: So one of the... I'm sure I'm not going to name them, but there was one chap who was on my MSc course who did really well on the course and then picked up a great job afterwards. And he was before doing the MSc. He was a used car salesman. I'm not kidding you. And he decided that that wasn't what he wanted to keep doing for the rest of his career, wanted to do something else. And he got into coding and I think he did, he started on his own. I think he got a private mentor to help him get going. And he built up his portfolio. And then he had an undergrad degree in journalism, so that enabled him to apply, to do the MSc course. And he worked really hard and he was just one of these guys he's... he was so passionate about it. So excited about it. Actually that was his problem in interviews. He got too carried away. He had to calm himself down, but he eventually landed the job that he was looking for. And he's got there in the end. There's plenty of examples out there of people who are really interested in it and want to get into it, but haven't necessarily done the higher education based skilling.
Bethany Rodgers-Rintoul: Definitely. So about having that passion and enthusiasm, isn't it? That can really help. So there's a question from Marvin as well. Marvin, I've seen your comments about the YouTube. If you're looking for recordings of things that we've done previously, if you just pop onto YouTube and search for endjin Solutions, you'll see recordings of the majority of all the events that we've done previously, and this one's going to be popped up probably first thing tomorrow as well. So just on YouTube search for endjin Solutions and you'll find lots.
So in terms of the question, then: "Are the only roles people look for in data science, machine learning teams still development and technical roles, or is the ecosystem of work now, so complex that roles like architects, project managers, product managers, and agile coaches and AI coaches," I suppose it's future gazing. Where the industry going, are they still involved in the kind of development tech roles or are they going into more advanced? Do we see where the industry going? What's your thoughts Delphine and Barry?
Barry Smart: I was just going to say Nude, where I worked for my summer internship was a great example of a startup where it was a multidisciplinary team. I wasn't working on my own. We had a product owner who was engaging with the end users. We had testers, we had an agile coach running the sprints and things like that. Data science is not just one role that, you've got to have people in other roles that kind of understand it. So if you've built up you've got wider skills to bring, to bear around any of these things or passion for those things, but in a data science context. Yeah. There's definitely roles out there in those kind of areas.
Bethany Rodgers-Rintoul: Hundred percent. I was actually talking about this yesterday with a client. 10 years ago software solutions were still a relatively new concept. There were something that were just coming into fruition and now, you can buy them, plug them in. You're done and dusted. Is that going to be the same in 10 years time with data science? Are we going to have a kind of one-off solution for data science problems that's along the same lines? What do you think of that? Delphine and Barry.
Barry Smart: Personally, yeah, the underlying capabilities might get to that point where they're a bit more plug and play, but I think you'll always need that translation layer. And it's just going to help the data scientists to do more in a faster manner. I think the role will always be there.
Bethany Rodgers-Rintoul: Absolutely. So there may be data science solutions in the future, but we'll always need to have the good communication people, the storytellers.
Barry Smart: Yes. Yeah. And it'll need configured and tested and just think of all the ethics. We haven't chatted about that at all, but there's even careers now emerging in understanding all the ethics around some of the stuff that's been developed, because data, unfortunately, we want to use it for good, but sometimes even inadvertently it can end up doing harm in some way. So there's even roles emerging around that as well. So that'll always be there, I think.
Bethany Rodgers-Rintoul: A hundred percent. There's a question from an anonymous attendee: "Do you think that most junior scientist, most junior data scientist positions have requirements that seem a little bit unrealistic for someone who's just about to start their careers?" I can see Delphine laughing there, or is this just really expected from someone who's about to graduate? For example, a junior data scientist position, graduate data scientist, position asking for one year plus experience and cloud services plus distributed computing systems. What's your thoughts on that one? Delphine. sorry,
Delphine Rabiller: I'm laughing because I remember when I was looking for my first job seeing that role that was advertised by a company was asking for somebody with experience in a programming language that was longer than when the programming language actually existed. I think that a lot of time with job descriptions are just put together randomly. And there's actually not much thought that's been put into what the key skills are supposed to be. A lot of people don't really know as well what a job is required as a data scientist. I went for an interview where I was told, "So we want to start using our data. We need to build the data structure we need... so everything actually needs to be done from scratch." And when I asked them how many people were going to be in the team, "oh, it's going to be just you." And how much time will I have? "You're going to have three months." It was, it's just completely unrealistic. A lot of people don't know what they're hiring for. That's why, what Barry and I were saying earlier is really important. Just ask questions, what do they want a data analyst for? What is it they want them to do? Just to see how realistic the job description is.
Barry Smart: And going with a few ideas of your own as well. Because they always appreciate that. Even if they're completely off track and you've misunderstood their business or what data they have, but just showing that creative mind as you go into the interview really pays dividends.
Bethany Rodgers-Rintoul: Absolutely. Absolutely. I think that's all that we've got time for at the moment, because I'm conscious we're wrapping things up at half past one. But genuinely thank you so much to Barry and Delphine for joining us today. And I'm sure that we all learned a good amount - very interesting insights from both of them. So thank you very much. Delphine and Barry for your interest and insights. Again, I'm sure we're all more than happy to connect with you on LinkedIn to discuss the conversation further. And if there's anything that we can do to help you moving forward, then we're more than happy to help. Everyone within the Scottish data science and technology community is honestly a lot friendlier than you would think. We've all been at this stage where we're starting off in an industry before and people like to give back. So that's it from me. Barry, Delphine, I don't know if you've got any final words to say yourself?
Barry Smart: No, we're all good.
Bethany Rodgers-Rintoul: Perfect. Thank you very much for joining. And I hope you have a good day.
Delphine Rabiller: Thank you very much, Bethany. Appreciate it.
Bethany Rodgers-Rintoul: Thanks. Thank you. Thank you.
Barry Smart: Thank you, Bethany. Thanks everyone for your interest as well. Thanks.
Bethany Rodgers-Rintoul: Bye. Thank you.