#221 [Radar Recap] The Future of Programming: Accelerating Coding Workflows with LLMs

Hi, I'm Maggie, VP of Curriculum at datacamp, and I know firsthand how overwhelming it can be to keep up with the rapid changes in data and AI. But staying ahead doesn't have to be hard. That's why our curriculum team has curated the best learning content to help you and your team stay ahead, covering everything from data and AI literacy for busy executives to cutting edge generative AI for developers, and everything in between. Whether you're upskilling yourself or your team, we make learning practical, engaging and flexible. It just works. Power your data and AI transformation journey on datacamp.com.

Hello.

Hello everyone. Welcome to the session. From Learning to Earning Navigating the AI Job Landscape. My name is Adal. I'll be your moderator for today's session. And it's no secret, given today's theme of radar given you at the moment that we're living in, that the AI job market is growing and there is increasing demand for AI roles. Today, however, there is also increasing competition for these roles. So we decided to do this session because we want to guide you through everything you need to know about finding a job in AI in 2024, from identifying emerging roles to essential skills to practical tips and strategies to enhance your job hunt success. And there are no better people to walk you through all of these today. And then Karen Tan, Dimitri Shapiro and Alex James. First, let me introduce Karen. Karen is a learning solutions architect here at datacamp and she is key in working with datacamp for business clients on building their skills transformation programs. She previously built Accenture's global Responsible AI practice where she championed ethical AI deployment for Fortune 100 clients. She has worked with the Alan Turing Institute on algorithmic fairness and advised the UK government on the ethics of AI guidelines. Karen was an adjunct professor for the ADA National College and an assess for Oxford University's online AI program. She ran a consultancy specializing in learning design for data and AI with clients such as Multiverse and 2Y Combinator startups. Karen, it's great to see you.

Hey, thanks for having me on here.

Awesome. And joining Karen is Alex James. He is the Chief AI Officer at Dataminer. Alex is a pioneer in AI with a rich history of building products used by millions across various industries including real time event detection, healthcare, self driving, cars, media and telecom. His career includes tenures at Yahoo, Telefonica, IBM, Fuji, Xerox, Siemens AT&T, Bell Labs and DigitalOcean. Quite a few companies here. On the resume, Alex is an early advocate for human centered AI, holding over 100 patents and publications in top Tier conferences and journals. He has been widely featured in the press including MIT Tech Review, CNBC, TechCrunch, and more. Alex also serves as a mentor at Endeavor and techstars and was an advisor to the Colombian government on AI policy. Alex, great to see you.

Thank you. Happy to be here.

Yeah. Alex has definitely managed his fair share of AI practitioners, so we're definitely going to lean in on the hiring manager insights. Last but not least is Dmitry Shapiro, CEO at MindStudio, a platform for model agnostic AI applications. Think about, we're talking about it behind the scenes. Think about it. The Zapier of AI, he was previously the CTO of MySpace and a product manager at Google. Dimitri is also a serial entrepreneur, having founded the web app development platform koji, acquired by Linktree and Veo Networks, an early YouTube competitor. He has extensive experience in building and managing engineering, product and AI teams and is a serial entrepreneur. He's definitely hired his fair share of AI practitioners as well. Dimitri, it's great to have you on.

Great to be here.

Awesome. So, just a few housekeeping notes. I think we've already mentioned this. There will be time for Q and A at the end, so make sure to use the Q and A feature before asking any questions, for asking any questions, and upvote your favorite question as well for any other messages you have or if you want to engage engage in the conversation. Use the chat feature if you want to network add people on LinkedIn. We highly recommend that you check out the LinkedIn group which is pinned on the chat above. If you post your LinkedIn profile on the chat, it will be automatically removed. So make sure that you join the LinkedIn group if you want to connect. And with that I think we can get started. AI is a very broad field with a variety of roles available. So what I want to first understand is maybe an overview of the different roles in AI today and what are the key responsibilities and focus areas for each. So maybe Alex, I'll start with you, given that, you know, you've managed so many different AI teams and data teams. So yeah, I'd love to see a lay of the land of the different roles available today.

It is very wide, therefore it's hard to put down in a single in one or two sentences. But in general you can have think of two extremes. One is sort of the PhD level, very deep expertise, and on the other extreme would be users of AI and that includes consumers where you're using applications, and then within that you can think of roles where you're building models from Scratch, whether they're small or big models where you're using APIs in terms of job titles or things that are more specific, I often think about it as research scientists. Those would be the ones with PhDs and fields like machine learning, computer vision, NLP, et cetera, et cetera. Then you have data scientists. Many of them would have either PhDs in fields like physics or master's in data science or a different degree. But they're more focused on experimentation, so their expertise is more horizontal. It's less deep and vertical, it's more horizontal. Then you have AI engineers. And one way to describe AI engineers is that they're very well versed in AI, but they do more engineering work in terms of making models more efficient, making them run faster, optimizing for inference, et cetera, et cetera. And then you have lots of things in between.

And, you know, we're going to talk about the nuances between those different kind of spectrums. Maybe Dmitry, given that you're building a platform for business users to use AI, how do you see business worlds evolving as a consequence of AI, given that we just covered that technical spectrum?

Yeah. So organizations themselves are changing quite dramatically of how they operate. And here when we're talking about AI, because again, AI is such a broad field, I think mostly we're Talking about generative AI, these things that have come out in the last two years with ChatGPT starting this mad run of things. And so organizations since then have been incorporating parts of generative AI into their processes. Right. Every organization sort of runs a bunch of processes, HR and marketing and sales and all of that. And these are these workflows that traditionally have been done by humans because humans were the only ones that can do it. But now with generative AI, organizations are able to take a bunch of these workflows and take parts of them and completely automate them, and so take the human out of the equation altogether, sort of changing how the organization itself operates for things that still require humans to interact with the organization. You know, oftentimes we're seeing now organizations building AIs that take the lead, and human is just there to supervise the AI and help AI make decisions when AI sort of is unable to make a choice. Roles that are still kind of driven by humans, like software development, for example, are certainly being transformed radically by generative AI. Software development, for those who have done it before, has been mostly, you know, let me try to figure out some architecture, and then oftentimes copy and paste things from Stack overflow, which is a dirty Little secret. Most things have been written on Stack overflowing and sort of copied and pasted and modified to create the software that we all use today. Well, traffic to stack overflow has collapsed because these AI copilots are extraordinarily good at being able to write large amounts of code. And so software development itself has been transformed radically and is going to continue to be transformed. Certainly anything that has to do with, you know, enumeration of cases or use cases. So things like marketing, where you have to like figure out all of these Personas and create content for all of this. Every role in an organization is being impacted by AI. But I guess I think the most important thing is that I'm certain that the organization of tomorrow looks really nothing like the organization of today and yesterday. Because of all of this automation, any organization is going to feel much more like a factory where you get sort, in a sense a conveyor belt of AI driving things. And humans are periodically involved to help make the process continue to go faster. Because in any organization the whole point is optimization for what? Productivity, efficiency, the ability to be nimble so you can catch opportunities and things like that. Then AI is great at doing all that.

So we're definitely going to unpack a lot of what are kind of the AI skills that business users will need in the future. And Alex, you mentioned kind of that technical spectrum of data scientists, AI engineers, even researchers. Dmitry, you hinted at software developers and business users. And Karen, I think your experience is really interesting here because you've sat as a program manager in AI in many various organizations. So maybe talk to us about that AI translator role in a lot of ways, or someone that sits between technical teams and business teams as well.

So I started my career in a consulting house called Accenture, and when I joined, I joined into the graduate scheme. But I knew that I wanted to be in the data and AI practice. Even though I came as someone without a technical background. Back then it was quite common that they would look for people who had statistics or econ or math or even computer science or engineering to align them to the data and AI practice. But because I had experience going to business school and seeing that data and AI was an up and coming trend that was not going to go down. I knew that it was going to be an area that regardless of what area of business you ended up in, it was going to become prevalent or widely used. And we see that now with generative AI, to Dimitri's point. So with the AI translator role, the way that I would look at it is There are AI translators who are more technical. So people come into the AI translator role, whether that be from an engineering background, moving more into product management in the AI space, or program management as it pertains to AI and technical projects, or there are people who come into it, like myself, having had a business school background, and then kind of approaching it from a perspective of, okay, so data and AI is useful for answering business questions, to guide strategy, to kind of grow and expand the business, and then kind of taking that mindset and thinking, okay, how do we translate these more technical questions into business value and vice versa? Like if we have a business question or we've got an area of opportunity that we want to explore, how do we translate those questions into ways that can use analytical techniques to then come up with ways to strategize and then implement those solutions or those hypotheses? And I think to the points of Alex and Dmitry earlier, so there's obviously still, and this isn't going to go anywhere anytime soon, we've still got the data scientists and the software engineers, and there are companies that are trying to build your generative AI software engineer or generative AI data scientists. But it's very much the case that, so we've traditionally been in this notion of the knowledge economy, where if you think about the white collar workers sitting at desks, we are ultimately being compensated for how quickly we're able to use our skillset and expertise and deliver the value that we've been put into as a role and responsibility. We're seeing the shift with generative AI and companies like Dimitri's, where we're moving from the knowledge economy into something someone coined, the allocation economy, wherein it's the case that it's no longer just the data scientists who are able to interact and get value from AI. It's also like the citizen person in finance or like your regular person in the strategy team or the operations team who are trying to use insights to guide decisions or to make important strategy directions that are now able to use notions like generative AI to come up with solutions. And so it's interesting because you mentioned my background is in product managing of AI and I've done other kind of more translation roles, being very close to technical people. But what we're seeing now is this notion where it's the domain experts who are leveraging generative AI and leveraging these tools to come up with solutions as opposed to the data person doing kind of like all that. So the allocation economy is an interesting thing that I'm seeing where we start to see people managing these AI assistants or these AI bots. And so that like translation and people managing skill is also becoming important in this space.

That's awesome. And we're definitely going to unpack a lot of that spectrum of roles that we were just discussing now, from business users, program managers, translators, to technical folks, AI engineers, deep learning practitioners, et cetera, and focusing on the technical Personas. First, I'll focus this maybe to Alex and Dmitry. To break into AI as a practitioner, someone who works on the nuts and bolts of AI, you need to have like a pretty strong technical foundation. So what are the technical skills anyone that wants to break into a technical role needs to have today? Alex, I'll start with you as well.

One thing that I wanted to clarify that I think is really important for everyone to understand is that AI is not just generative AI, a hundred percent. You know, they think of ChatGPT as AI and they think that's AI, but it's not. There are two categories of AI. There's predictive AI, there's generative AI. Predictive AI is all about producing labels or labeling things. One example of this might be email, right? So you get email and your email system will classify emails into spam or not spam. That's predictive AI. When Netflix recommends a movie for you that is predictive AI, it's just saying you're likely to like this movie, not this movie, et cetera, et cetera. That's predictive. Predictive has been around for decades. It's been applied across advertising, search, in many, many industries. We've all used it for a very long time. So it's been around for a really long time. That's predicted. Generative AI, as the name suggests, is AI that generates text, images, video, audio in a chatbot, it generates the answer to a question, or the next word in a sentence, or the next paragraph in a book, et cetera, et cetera. So it's important to understand that these are two categories. And something like ChatGPT is an application that uses underlying AI models. So you should distinguish the application, ChatGPT copilot Gemini, the application midjourney, the application, and the actual models. So anyway, really, really critical for anyone who wants to get into AI, whether it's on the technical front or on the applied side to understand those very basic concepts. Because if you don't, you're going to stumble into lots of confusion. Now, in terms of the skills, it really depends on where you fall into that whole range. And we talked Dimitri and Karen added some more roles. So there are definitely many on the application layer where you're a practitioner in some domain, whether it's nursing education, whatever it is. And there are a bunch of tools out there that you can use. Some of them include one that I mentioned already, ChatGPT and the CO pilots, but they're industry specific tools. And so for those, the skills you need are going to depend on the tasks that you perform. And so I think for people that want to get into AI within their domain, the skills you need are going to be very specific to your specific tasks and field. And what you want to do there is see what tools are out there for. If you're an attorney, you know what are the tools that attorneys are using? Are there any companies that are producing AI for law that you can leverage or for marketing? And then for the other ones, for the more technical ones, what I usually suggest is to look at lots of job opening descriptions and that will give you a sense of what people are looking for. And that's going to depend a lot on the type of company, the stage the company is at. Some will say we require a PhD and unless you want to do that, then you would say, well, that's not really where I'm going to fit in. There may be another role where it doesn't say that and it just gives you a bunch of tools and say you should know Kubernetes, Pytorch, you should know how to fine tune LLMs, whatever. It's very hard to give you a formula because I think it really has to be very specific and each person has to do their own research and see where they fit. And then once they see where they fit, see what's missing and then you go and you try to learn about those things that you're missing and then.

Maybe taking that a step further and I'll ask this question to Dimitri, because your company, your team is actively building generative AI solutions. When you look at the stack that you use, Dmitry, what do you think is a canonical set of technologies and tools that people want to work in? Generative AI today need to learn stuff that comes to mind are like hugging face, LangChain, et cetera. So I'd love to get your view here.

Maybe I can add to. Alex is a little bit first and zooming out when we talk about stacks, I encourage folks to think of AI like anything else in tech as a stack. We're sort of at the bottom most layer. You've got infrastructure and so there, there are lots of interesting opportunities and Work to be done on building GPUs and building efficient data centers and lots of things that are done on the infrastructure layer. And so you could go deep into that. And if that's what turns you on, there are tons of opportunities that's only going to get. We will require more and more optimization of the infrastructure layer. And there's a tremendous amount of money that's flowing in there. Those people are constantly in need of new people to come and help build that infrastructure. On top of the infrastructure layer sits the intelligence layer, and in the intelligence layer there are all the different models. And those could be language models and image models and code and video and sort of countless things that are the intelligence layer there too. There's obviously lots of companies, lots of opportunities, and if that's what turns you on, you should do that. And above that sits the application layer. And I actually believe that for most people, I suspect on this conference here that that's probably going to be the place where you will find the things that you're looking for, because that's sort of the. How do all of the things that we do as humans benefit from all of this infrastructure and intelligence? You know, so what we've got all these models. What are you going to do with them? Well, every organization, every company that exists now understands that they need to figure out how to become more efficient because this next stage that we're in is all about, as I mentioned, sort of enumerated efficiency, productivity, intelligence, et cetera. And so in the application layer, the skills you really need aren't actually technical skills per se. They are skills that are much more akin to product management. The ability to look at something and understand a process and be able to break it up into pieces and then be able to articulate primarily in human language, what is this thing, how does it work? And if you were sort of going to take this piece of the process and teach that to some new employee that came in, what guidebook would you give them of like, here are the steps to do your job. As far as your sort of original question of like, what's the tech stack for building this, unfortunately it seems very wishy washy and shifting dramatically day by day. And so, you know, you have LangChain and llama index and vector databases. More and more people are realizing that today, sort of a pattern called retrieval, augmented generation and people may have heard about is sort of a stop gap. I actually don't think that's sort of the long term solution to any of this stuff. But at this moment is critical for being able to build applications that are robust. This is sort of at the lower level and then it's sort of standard tech stack. Like if you're building applications that are used by humans, you might need to build screens, input output forms and screens. If you're building applications that are headless, you need to build API endpoints and these things expose those. So there's sort of no unique magic there. Basic software development in doing that. You know, Mind Studio, we try to package all of this together so you never have to think about any of this stuff. Sometimes people describe mine Studio sort of like non technical LangChain or like LangChain for developers. So business users can build their own applications without understanding anything like LangChain or, or Llama Index.

So we kind of covered here a bit the technical skills. Especially like the stack needed to break into more of a technical practitioner role. But I think as well, you know, you hinted that this as well, Dmitry. And I think, Karen, this is also very relevant to what we've discussed before is understanding if you're going to be able to communicate in human language how a process works in AI, right. And being able to kind of really operate within the organization using AI as maybe a business user. Right. What you need to have is really strong storytelling communication skills. And I think this is whether you work in data or in AI communication, storytelling skills are really essential. So maybe how can anyone in this, in this field develop strong data storytelling and communication skills? And can you develop this skill even? So, Karen, I'll start with you.

I think that's a really good question because when you think about the space of data and AI, the fundamentals, it's like a very technical area. And so oftentimes people who are maybe stronger in non technical subjects feel like they can't enter the space or break into this area because they don't have data science skills or they can't do the analysis themselves. Obviously those are very important skills in the space because otherwise you wouldn't be able to have the answers that you're looking for. But there are people who work closely with those technical talent or the technical talent themselves where it's to your point, Adele, very important that you're able to tell a story or to narrate what the data is saying. Especially because oftentimes when the analysis is very deep, the audience that you're trying to communicate that to, to help make decisions may not have the context or the depth of expertise in going deep with the numbers and the data. So the ability to tell A good narrative is very important and to be able to go from insight into action is like super important, especially making very high level decisions. Some ways to think about your question, can you develop these skills or can you improve on it? Ways to, to increase your ability in that area. You can think out of the box things like doing a theater class or if next time there's an opportunity to present at lunch and learn a brown bag or some presentation at your company or the place you work, or even at university, if you're a student, make the most of those opportunities and take those as opportunities to learn and develop further. And then obviously seeking feedback from those around you is a really good way to improve yourself in that area.

That's really great. And Alex, Dimitri, I'll also push that question to you, Alex. How have you improved your own storytelling skills throughout the, throughout the years and how do you improve your team's data storytelling skills?

Yeah, so I think one of the key things in tech, and this applies to AI and not AI, is communication and sort of understanding the business value of what you're doing, etc. Etc. So I would say that it's not just AI, however, there's a key difference with AI work. And the key difference with AI work is that it's often non deterministic. And what that means is that the results of what you're building depend a lot on the actual inputs. So if you build a model, you test it out, you train it, once you put it in production, it may behave slightly different from what you expected because it's seeing new inputs that it had not seen before. And what that means is that before you put it out there in production, before you build it, you need to have a better understanding of what are the key metrics, what are the business goals, how do you think it's going to perform, what the risks are so that you deploy it responsibly and ethically. And a lot of this should be communicated to your peers and upwards in your organization. So that's where the communication skills become more critical than in traditional software engineering. Because you have to pay more attention to these subtle shifts in performance and the risks that deploying AI may have, whether it's a chatbot or it's a predictive AI model that it's making predictions, it can have pretty significant impact. And that includes the cost of errors. Right. If you have an AI model that is meant to predict whether somebody may have cancer or not, a false positive versus a false negative are two very different types of errors. If a patient Comes in and the AI model says, you don't have cancer, Go home, it's all fine, come back in 10 years, and it turns out they do have cancer. The cost of that error is very, very high. As opposed to the AI makes some mistakes and says, maybe you do have cancer, go do a second test and you don't have it, that error is a lot lower. So communicating these is critical in the AI space. And I think for every project you work on, what you want to do to improve your skills is frame it that way. Like, think of the questions that people ask, think about the errors, think about the implications. And I often have people in my teams present progress and the goals of their project, so they practice communicating.

That's really great. And Dimitri, you know, you're expert communicator, maybe shared some of your secrets with us.

Until I was 22 years old, I had practically no communication skills whatsoever. I couldn't even look my friends in the eyes when speaking to them. I was a total nerd, like the worst. And I think one of the smartest things I did was I got a sales job selling phone systems to businesses. And when I told my friends in college that, hey guys, I got a job, I'm going to sell phone systems, they're like, you're kidding. If you're going to sell, you got to talk to people. You can't even talk to us. And I was like, yeah, I have to change. And I very much encourage anyone to do that, even though most people are really scared of getting a sales job. But I actually think that is the best training. Sort of the answer to your question, how do you develop these skills? You interact with other humans. And when you do that, you start to realize that communication is really, really hard. Specifically nuanced, disambiguated communication is really hard. And so you might think you're communicating with other party, but for whatever reason, they're not buying what you're selling. Well, what is that? Maybe it's your product that sucks, but probably not because you're going to be smart to choose a good product. Maybe it's because you're not sort of meeting them where they are, understanding their perspective and then being able to lead them to where they need to be to understand that your product is awesome and for what it is that they need. And so this ability to one, analyze a situation, analyze a need, again, this is like an important analytical skill. It's not technical, it's just sort of the ability to look at something, say, how does it work? How does it need to work. That's important. And then again, that ability to articulate in a nuanced, disambiguated way the situation and the need. What do you want, in a sense, the AI to do? I encourage people that are building AI applications again, to think of AI as being this new employee that they've brought into their company. This employee can do anything. They're brilliant. But what do you need them to do? Well, you can say, I need you to go and run this process. An employee says, great, how do you want me to run the process? Okay, well, that's a hard question. How do you articulate to them in a nuanced way what the edges are, what they should do or what they shouldn't do? And so you can absolutely learn that. I'm an example of that and I've seen certainly many other people do it. You can do it now by again, sales is great, or you can just practice with your friends of try to explain something to them and then get them to say it back to you. And what you might realize what they say back to you is not exactly what you said to them. And then you got to figure out how to add some more words or sort of refactor your communication. So tuning for when they say it back to you that it's exactly what it is that you meant. And that's just a function of practice.

I love all of these advice. All of you are expert communicators and great storytellers. So it's really great to see how these different advices fit together. And maybe my last question. It's definitely no secret that we are in the middle of a hype cycle when it comes to AI. And even when thinking about the skills agenda and the roles, we've often heard a lot about the prompt engineering role skill set over the past couple of years now. I am personally am very bearish on the idea of a prompt engineer job in the future, but I am very bullish on the idea of a prompt engineering skill set. From your point of view, is prompt engineering really a role to look out for or just a skill set? Maybe? Alex, I'll start with you.

I think it's a great way to understand what the models do. So I encourage people to do it. Whether it's going to become sort of a permanent role, I think remains to be seen. I'm kind of with you in that sense. I think there will be applications on top of the models that will make things easier and that will be more specific. But I think what it does Today is that it helps you understand what the models can and can't do. And for certain tasks, that is the way we're going to interact with some of these systems. So it's worth for everyone to learn at least some basics of prompt engineering.

That's awesome. I couldn't agree more. Everyone needs to learn some prompt engineering. But I don't think there's going to be professional prompt engineers. I'll clarify that point. Maybe I need to improve my communication skills. Dmitry, thoughts on the prompt engineering role skill set?

There is no such thing as a prompt engineering role. It will not exist. It doesn't make any sense. Prompt engineering is a fancy word for this thing that I was describing earlier. The ability to express what it is that the AI needs to do to express a process in the form of architecting the workflow of the process and creating constraints on the process. I mean, we can call that prompt engineering. But this role of like I'm some expert AI whisperer, that's just silly.

Completely agree here. And maybe Karen, let's get final outcome.

Yeah, definitely on the same page as everyone. If you think about like these models that people calling prompt engineering roles, ultimately it's like this library of the Internet, the public Internet. And so if anyone would be a prompt engineer, they'd be kind of querying the public Internet. But the most valuable person to do that would be a domain expert. So it's more likely the case that domain expert would upskill in prompt engineering to get the best outcome. So final words on that.

Okay, awesome. So I'm going to stop asking questions. I'm going to take questions from the audience. We have around 10 minutes or so. First one, I'm going to show on stage. I think a question everyone wants to wants to see here, what strategies can help someone stand out when applying for AI positions in a competitive job market. So maybe. Alex, I'll start with you.

Yeah, I think you need to have the skills that comes first. You can't really stand out if you don't have the skills. You're not going to succeed if you try to fake it. I do think working on projects and building stuff is the number one way of showing your value, whether it's independent projects or it's projects you've worked on at your current company. If your role is not currently AI and you can somehow participate in an AI project within your company, that would be a good way to move forward. But if not, spend your weekends, do it on your own. I would also include taking some online courses, getting certifications of that sort. And there it's a matter of strategy. If you're at the point where you're like 100% sure that's what you want to do, and you decided you want to do a master's in data science or in a related field, or do a PhD or whatever, that's a longer term position, but that will help you stand out in the long term as well.

Great. Dimitri, from your point of view, how do you stand out as an AI practitioner?

Yeah, I don't think that this is sort of exclusive to AI at all. I think it's like anything else because again, you're interviewing with a human and the human is deciding whether they want to hire you for their company. So it's a function of how well can you communicate. Obviously having skills as sort of table stakes, you have to be able to have the skills to do the job.

Check.

Then it's a function of culture fit. It's a function of how excited are you about what the company is doing, how much can you articulate the value you can add? You know, I've hired many people over the years. I always look for people who just need to get this job. They love the work, that's what they want to do. And you can see it in their eyes. You can see it in the way they speak, you can see it in the questions that they ask. When I coach people to actually quite often to interview, I suggest to them that they take control of the interview, that they show up in the sense they are interviewing the interviewer and those are the people that you're looking for because they show up and they understand what it is that the organization is trying to do. They understand what they want to do, and they're asking all the right questions. And that becomes a conversation. And that's the way you stand out. It has nothing to do really sort of with AI itself. It has much more to do with human to human communication.

I couldn't agree more. Karen, what's your number one advice for standing out?

My number one advice for standing out is to be an unforgiving and boisterous hustler. So obviously we're in a period where there's a lot of layoffs happening in the tech sector space. And it's a very demanding space. More and more people are getting masters. So to have a master's is like becoming commoditized. So how do you kind of stand out from that crowd? I often suggest to people, and I do this myself when I'm looking for jobs or opportunities. Or even when I ran my own consultancy, I often look out for roles on LinkedIn or even when people post things, and if they put a number on there, I'll just call them and be like, hey, I saw this. Or if I see that a company's hiring, I'll click in to look at who the hiring manager is. And if it's a small company, I will just ping the CEO with my CV and say, hey, I saw this role. And I've gotten so many consulting gigs through my own company and then also in other capacities in this way. So I highly recommend doing that process, because if you're a good candidate for that role, maybe there are hundreds, but you're feeding yourself on a silver platter to the person who's out there actively searching for you. So that's the best way.

Couldn't agree more. Definitely. Being a Bolster's hustler, I'm going to get that tattooed. Yeah, maybe. One additional question as well here that we have. I think this is also kind of relevant to what we just discussed. AI is applicable to a lot of business fields. You know, I think, Dimitri, you laid that case really well. And as someone from an arts degree, how do I break into AI? And then someone asks, I'm new to AI. How do I break into AI? So I'm going to combine these questions into one. So if you're someone from an unconventional, let's say, humanities degree, right, and you want to break into AI today, how do you get started? Dimitri, I'll start with you.

You actually think that's the best degree to have? I have five kids, and periodically people ask me, well, like, what are you going to suggest your kids study in in college? And my answer is, I'm going to try not to push them in one way or another. But if. If they sort of forced me to push them of what they should study, I would say liberal arts and anything having to do with communication, I think those are the sort of fundamental skills that you need. So if you're an artist, that's fantastic. You should be able to do it. If you're new to AI, everybody's new to AI, meaning everybody's new to generative AI. And so that doesn't really matter. I think throughout the session, we've already talked about what you can do. Learn to be able to communicate well, learn to be able to analyze situations and processes, reach out to organizations that you want to work for. I very much believe that there's opportunities in every organization for people to be able to show up at an organization and create their own job to reach out to some company that you want to work for and say, this is what I would like to do. Is there an opportunity to do that? And oftentimes your skill set they might not be looking for, but when you show up and you articulate to them what you want to do, they might realize that you've got the perfect skill set to do this thing you're proposing. And that thing you're proposing would actually be valuable to the company. So create your own jobs. And so I wouldn't be at all discouraged by I'm new or I've got an art degree. Again, an art degree is actually great. Just go out and engage.

That's really great insight. And Alex, maybe how do you get started? Like, if you want to transition into AI today from an unconventional background, what would be your advice on how to get started?

I think there are two questions and I keep seeing them in the chat. One is, I'm a teacher. How do I get into AI? So the question there is, are you going to leave your teaching job and become an AI practitioner of some sort? That's one category. And for that one, I think you need to, as I said before, look at job openings, get some basic training, understand the concepts, and then you see how deep you want to get. There are different layers and many different opportunities. The second one is I'm a teacher or a nurse practitioner or whatever it is, and I wanted to get into AI, but that doesn't mean I'm going to stop being a teacher or a nurse practitioner. I just want to use the tools in my career. And I see most of the people are going to fall into that category. And so for that one, it's what I said before. Try to understand the basic concepts and then use the tools that are available. And you can start with the basic consumer tools like the co pilots, chatgpts, et cetera, to understand what's out there. But then look at specific tools within your domain, like what are the AI tools for teaching, for nursing, for law, for whatever that is. And there's no question that these AI tools will become more prominent in all of these fields. So the faster you get to use them, the bigger your advantage will be against other people that are competing with you for similar jobs. Because at the end of the day they're going to say, well, if I have a lawyer that knows how to use AI tools really well and one that doesn't, I'm going to pick the one that knows how to use the.

Tools If I could maybe add to that from sort of a different angle. As I was saying earlier, every organization you know, in the world needs to figure out how to leverage AI to become more autonomous, more productive, etc. Most of those organizations, vast majority of them, have no skill set to be able to do that. And so there's this great opportunity for people who know how to do that, who learn how to do that, to approach those organizations and say, I do that. I do AI driven business process automation. I do this typically known as digital transformation. I can come to your organization as a consultant or as an employee, look at your workflows, automate a bunch of things, create custom applications, etc. So that's a kind of the biggest need, I think, that exists now and the easiest one in a sense to get into. There's like tens of thousands of people that have learned to do it using Mind Studio. So you can certainly do that or sort of whatever other stack you want to use to be able to do that. I think that's actually the biggest opportunity. Like it's not, can I get into OpenAI or anthropic? Those positions are going to be limited just by definition. There's pretty much unlimited demand for a digital transformation leveraging AI inside of any organization, including schools. So if you're a teacher, you can take your school and make it radically more efficient by learning how to build AI applications for your school so you don't have to quit your job. But you, you know, we've seen, actually I know of at least a half a dozen people who have learned to use mindstudio, an HR manager, a project manager, et cetera, and have now been promoted in the organization as being like the head of AI automation or things like that, those kinds of roles. So there's nothing but opportunity out there.

I couldn't agree more. I also have a very personal experience of that where I augmented my own role in a previous organization, showing me into a promotion. I'm going to ask one last question and it's going to be a tough one because it's a prediction question. So I need a one minute prediction from the three of you, which is what do you think the AI job market will look like 10 years from now? Karen, this is, I'll start with you.

I think that a lot of what we see now is still going to be there. What ratio of it will be to be determined. But to the point, what we've talked about, there's a lot of people that the startup space is exploding with companies being funded Trying to figure out ways to build value, disrupt different industries using generative AI, let alone the more traditional predictive AI, which is also a space that's also funded. So I think that a lot of what we see now will still be there ratio tbd. But then in terms of, I would almost flip the question to not what would the AI job market be like in 10 years, but more the case that generative AI will kind of infuse into every job role. So it almost is like AI will just become part of everything. Whether that's very visibly a technical role or whether that's rather than using Google Search, maybe someone's using Perplexity or a GPT that they've built.

Those are my thoughts. It's like the Microsoft officer of AI in a lot of ways. Office became quite predominant in the 80s or something along those lines. Dmitry, I'll ask that question to you now.

Yeah, I agree with Karen. I think the question is better asked in a different way. If you want to build sort of core technologies of AI again in the infrastructure and intelligence layer that I think generally is going to continue to evolve. Hard to predict how it's going to change over the next 10 years. I think generally it's going to be relatively similar. I think for the next layer, the application layer of how AI transforms organizations. I think that too is pretty clear in what I've kind of been saying throughout this panel, that the vector is efficiency, productivity, automation. And so organizations are going to strive to become radically more automated and more efficient. And so the, the need for humans to participate in that certainly is going to change. It's going to vary organization by organization. But I think the jobs are going to be quite different than they are today because again, generative AI, this generation's generative AI now can already do incredible things that humans are doing manually.

Wow.

Ten years from now, especially as things are accelerating exponentially in capabilities, I think it's hard to predict.

That's wonderful. And Alex, I'll let you finish us off.

I want to repeat again what I said earlier, that there's generative and predictive AI. Most of the value today and historically has been in predictive AI. Generative AI. We're just beginning to see what that looks like. I think like, you know, if we think 10 or 20 years ago what mobile phones were going to do, smartphones, we're going to do in the future, we're going to see many, many new applications we had not thought of. And in the same way, everyone's going to be using AI. We're already seeing that very, very quickly. It's creeping into every consumer and every business application. So it'll become sort of a de facto part of what we do on a daily basis. It'll impact every single industry. So when you think about the job market, I think it'll still be a wide range of areas and skills that will be needed. And at a very minimum, everyone will need to be able to use these tools efficiently. And in terms of what the core technology is going to look like, I think that remains to be seen because it's moving very, very quickly. But I think there will be opportunities across the board. And that's why I think it's important for everyone to get started as early as possible. And especially understanding the concepts, the capabilities, limitations, and in using it, you know, you can't learn this stuff by reading about it. You can only learn by actually using it. And the more you use it, the more capable you will be in leveraging these technologies as we go forward.

And I think this is a great way to end today's session. I want to say a huge thank you to Dmitry, Alex, Karen, I want to say a huge thank you to everyone who attended this session. It's truly a pleasure, everyone, to be able and a privilege to be able to be hosting these sessions. I want to say huge thank you for everyone. And again, one final thank you for all of our speakers today. I really, really appreciate.