Unleashing the Power of AI: Elevating Career Readiness

you know well good morning and welcome everyone to our first backpack to briefcase of the Year we're so excited to be able to start the semester off by talking about how students can use AI to elevate their career readiness share with us as our own resident expert and Guru Dr Sharad Jones thank you so much for taking time to be with us today and to cover this important topic by way of brief introduction for myself my name is Valerie Reese and I'm a career coach here at the Huntsman School of Business and the backpack to briefcase series was created as it means to talk about all things career preparation for our students next I'll introduce Dr Sharad Jones let him take the time he needs to cover this important topic then we'll have a few questions followed by a few brief announcements so without any further Ado I'll introduce Dr Sharad Jones he's a professional practice assistant professor of data analytics and information systems in the Huntsman School of Business at Utah State University he earned his PhD in statistics in 2021 from USU where his res where his dissertation focus on conditional generative adversial adversarial networks for multimodal scene generation his current research interests are in multimodal neural networks ethical Ai and Transformers sharad's dissertation was done in collaboration with maxar Technologies which is a space technology company where he was employed as a data scientist from 2018 to 2021 prior to his PhD he worked as a data scientist for carvana an online used car retailer from 2014 to 2017. following his PhD he worked as a senior data scientist for strive Works a government contractor focused on developing an operational data science platform sherrod's wife Carly Fox is a lecturer in the data analytics Department as well and they met in Arizona and they lived in Cache Valley since 2017. some of sharad's personal interests include cycling both mountain and Road and baking and land photography so again we're so excited to have you here I will turn the rest of the time over to you and we'll have questions afterward perfect well thank you for the introduction it's very polite um yeah hopefully that gave you some context on my background on where I kind of fall into this and I guess why I have or should have any opinion on this space of AI so if you you know caught that in my dissertation topic I was you know working in this space of generative networks which was kind of what we you know as AI today right or at least what a lot of people are thinking about in the context of AI machine learning goes much broader than that and we'll talk about that here in detail but just understand that uh you know this field is moving fast and so you know we'll try to demystify some of it before we dive into kind of concrete steps you can take to start preparing for and understanding how you'll fit into this world um or out in the Working World with this new technology that we've kind of all had now so either way jumping into it this is a slide I try to show as often as I can I I think it's really really critical again to understand that what we're talking about here is not the AI that you've seen in movies um where you know the robots are controlling everything and you've got to worry about that like that is a completely different thing that's usually referred to as artificial general intelligence or even artificial super intelligence we do not have that today um there's opinions and debates on whether we'll get there whether it's even possible whether it's something we can Define or not and so I just try to keep that as an irrelevant topic right so that's why the artificial general intelligence intelligence there is kind of grayed out it'll probably include some of what we have today but it really is a separate entity so instead I want you to think about this as kind of narrow Intelligence being defined as you know being able to focus on or do a single topic really really well potentially better than a human can do it but it is narrowly focused it might seem like it can do multiple things and it might be able to integrate multiple tools that can each do their own thing but they are still kind of siled in that sense and so under that umbrella of narrow intelligence you have the field of machine learning and that is actually really what we're talking about with most of these tools today um there are other things that have been done historically in the world of AI you know some rule-based systems but it's machine learning that really has kind of taken the World by storm um and AI tends to be more used as a buzzword and that's okay we definitely should have those and it's good to kind of Encompass ideas under a single umbrella but I do want to make it clear here that we're talking machine learning and really machine learning is just pattern recognition uh the models like GPT for example right that underpins uh chat GPT that you guys all probably use at this point uh that is a model that's designed to try to predict the next word in the sequence well the way it predicts the next word in a sequence is by looking at patterns of words and sequences so there's some sort of input and those inputs have their own Associated outputs and so given those we're now just going to use statistics right some cool mathematical tricks to say well what are common patterns that appear right given some input sentence what typically leads to this next word in the output sentence and if I give it trillions and trillions of examples I eventually start to learn some sort of mapping between inputs and outputs and once I have that then I can kind of deploy it to be used in the wild right and that's really what chat gbt and all these other tools are is just something to be deployed or a better statement something to make those predictions based off of you know countless prior learned examples and I'll be clear to you again we're going to be talking mostly on large language models these models like Chachi BT that are doing word prediction but there's so so much other stuff that falls under AI machine learning really that we cover in all of our courses so please do take some I'll make Shameless plugs for our courses later on uh but it's really important to understand that anything where I'm making predictions based off of some sort of Prior inputs or trying to learn some sort of relationships or patterns in the data anything like that that's typically going to fall into this umbrella machine learning so you know even things like Netflix recommending the next best movie to you that's still under machine learning it's still using the same techniques oftentimes the exact same algorithms as you have in tools like chat gbt and you know this is just kind of highlighting this again and we see it everywhere right we see it with uh self-driving cars you know generative images a lot of my work again was in satellites imagery and specifically in generating images from satellite imagery so it's still all kind of touching under these same buckets Tick Tock is a really great example of it if anybody's used Tick Tock they know how kind of scary good that algorithm can be sometimes that's machine learning that's pattern recognition that's given a bunch of examples and uncovering patterns from that okay and so that's just what I'm trying to say here is that these are things you're going to see everywhere and honestly really any industry you plan to go into any industry you plan to work in is either already heavily using these tools or quickly adapting to and using this um many many companies I've talked to in recent you know months really uh have I don't want to say pivoted their businesses to focusing on machine learning but they have definitely started to at least ask that question of what does AI ml mean in the context of our business and so even if you're going into a marketing role uh you know entrepreneur or an entrepreneurial role a HR role they're all starting to be impacted by these topics and so it's really really critical to at least understand what this landscape looks like and then also today we're going to focus a bit more on is how you can leverage it then right how you can use these same Tools in your search right in your career search so that's our hope for this and we'll kind of get into it now I have a few slides and then like Valerie said a lot a few questions at the end but backtracking for a second because I can't help myself but we give a little history lesson as a professor uh it's kind of important also to recognize how fast this is changing and this is coming from somebody who has been in this field and it has already been or always been changing very very fast so again I started in the world of data science back in 2014 we had started to understand or leverage neural networks at scale mostly in imagery that was where they were starting to show a lot of success um and that you've probably heard this term neural networks and so really all that is is a very generalized form of linear regression so if anybody's seen y equals MX plus b we learned some sort of linear relationship between variables neural networks are kind of a generalization of that and saying well we can learn kind of these nested layers or multiple layers of interactions and they don't even have to be linear right so that's just a way to uncover patterns in the data and again 2014 was the first time this kind of became possible or I'll say maybe the 2012-ish on images but text came much later so you know it wasn't until about 2018 2019 when we started to see early large language models and these were built off of this transform architecture which we're not going to get into here but from there it kind of just snowballed very very quickly and around that 2020 2021 time frame was when some of the larger gbt models came out and then as we're all aware kind of late 2022 was Chachi BT and now we're kind of you know in this world that we have today and so these are kind of the three bullet points I want you to think about in the context of the job you're going into and again I'm you guys are all still in this phase of trying to find that next career opportunity but it's important to recognize well how are these tools changing are changing our career options because as they change them the opportunities that we have change for example I did not go to school for statistics or data science or at least my undergraduate degree it was in physics I was doing an entirely different thing but this world of data science was starting to Boom it around that you know mid-2010s area era and recognizing this changing opportunity I pivoted out of that and said well physics is really problem solving and using information and data and so I think I can start to tackle this but I didn't even really know the title uh you know it existed prior to that but I didn't really understand that job when I was starting to go into it and so being open to these new ideas and newly emerging Fields is very very critical and understanding what's kind of coming up can help Define somebody that and so first thinking about this automation of tasks um we've always had automation of tasks right that's like think about a production line they've had robots come in and you know robotic arms move things around quicker and can do parts of this more efficiently and so we've always done we're taking steps to increase you know the automation of some sort of process some sort of task right because that means we can be more efficient produce more increase Goods or you know lower costs it has the positive outcomes but that does also come at the cost of you know usually taking away the need for a human in some capacity in that space and what large language models have provided us is the opportunity to automate very abstract tasks right so again if we think about I am a professor here I teach data science but prior to that working as a data scientist I wrote a lot of code to you know performs or develop some sort of machine learning model to train on some sort of data set I can pretty much take most of the work I've done prior to that and automate this via large English models I took my dissertation actually for example um I worked on my distribution probably about a year and a half two years straight you know writing the code for this and you know aside from the fact that I kind of knew what I needed to answer using large language models using chat EBT I was able to recreate my dissertation work and probably the span of like a day or two um it was kind of alarming to me you know a little bit of that moment of like what am I supposed to be doing here but it was a chance to reflect on you know maybe it's not a bad thing that I have been able to automate you know these really abstract tasks but they are tasks that we can Define decently well and so now I can just quickly iterate through them and so thinking about what parts of your job that you're interested in might start to be automated away will help you understand what to focus on in your learnings right in your time here at USU uh but again with every bit of automation typically comes some sort of new job creation so we don't usually see uh historically a lease that when a new technology is created that the amount of unemployment increases it's usually actually that more people tend to get jobs now some people do lose their jobs it's just that it's offset by other things in the economy so there is this aspect of certain people in the labor market do get you know kind of missed in this bucket but there are new job opportunities and so thinking now in the context of AI I mean I can't tell you at all what your field will for sure have what new jobs will start to pop up in that field but I can at least say that the things that were repetitive will be gone right those will likely be removed or automated away in some capacity there will be products built and so we'll need people to build those tools right so there's that one aspect of it only people that know how to effectively use those tools as well um you know even if you think about something like a computer right the development of the personal computer automated a lot of monotonous tasks but now all of a sudden we have to learn how to use computers effectively and be very efficient on it and learn how to use spreadsheets learn how to create PowerPoint presentation all these kind of things we had to learn outside of this that we're leveraging the tool it's going to be the exact same thing with all of our AI tools all of our machine learning tools we need to learn how to effectively use them and kind of be a part of that system it's not going to remove humans they will almost certainly all be assistance and so being able to very efficiently and effectively leverage those assistants is going to be absolutely critical because it's going to hit us with that third point right it is going to increase productivity insanely um again I've only had these two as long as you all have had them but the amount of time I've saved by being able to you know quickly outline outline plans of you know even things like course development or again writing code it is insanely good at writing code so I can go in and say well I have this big walk the idea I don't really know where to start on it maybe you know generate me an outline of how to proceed on this give me a skeleton you know framework or you know kind of wireframe of the code I need to use to build this application and then I'm going to iterate on this right I'm going to move faster now and I'm going to try things and break things but that increased productivity gave me an opportunity now to do so so much more than I've ever been able to do and so that's really where it is now it's kind of your choice what to do with that Spare Time some people I've seen are taking this as saying great part of my job is now you know easily done by chatgpt my boss doesn't know and so I'm just going to kind of you know chill here for the next seven hours of the day but I think the more effective thing to do is to think about okay if it can do my job in a fraction of the time that I maybe need to think about what are the things that I actually feel are most important to me to my job right so to now pivot a little bit and think about what this means in the job search process we've been talking about how it's going to change the job maybe but now how does this change kind of the search process and so here are some examples of ways I've actually been using it very heavily so for or I guess we'll kind of jump to the last bullet points because you can see here in this image on the right that way you can look at that but either way uh that is an example of uh chat gbt or sorry this specifically Claude so Claude is a different model built by a company called anthropic but I gave them it allows you to upload documents so I uploaded two documents I uploaded my resume my personal current resume as well as an example job description I think this was for a research scientist at open AI so the company that built uh chat gbt and so I said you know attaches my current resume and I'm posting a job I'm interested in I know that my skills are relevant to there but I don't know that my resume is necessarily tailored to that job because right now I haven't been focused on applying for research scientist type jobs and so I don't really have a resume that reflects that so maybe using this tool to help me refine and deter and Target my resume more towards the ones that the jobs I'm actually interested in as opposed to just kind of having this generic resume that maybe everybody has because again in the era of AI and the era of machine learning these companies are also using tools to really quickly and efficiently scan resumes so the more generic your resume is and I understand that you know early in career my resume was very generic and it to some extent still is amongst data scientists I need to say something or highlight something that really shows my unique story or my unique ability to contribute to that space so I've used it heavily for these kind of things you can use it to help you write cover letters it's outstanding at that you can say hey here's my resume here's the job description I don't want you to necessarily write my whole cover letter but give me some bullet points to highlight to pull out that can start to capture my story and how it's relevant to them you can very much ask it to write your whole um cover letter but I will warn you that the writing from these is never perfect you know it's a little bit robotic still and you may it will get better and it will likely get to the point that it is believable but understand that as Things become uh or better statement is these tools become more able to write more content the details of human writing and the details of human understanding will start to matter more right people will start to recognize that everybody writes some generic long uh perfectly tailored cover letter and that's because they're all using these tools as well so you'll still kind of have to put your individual Flair on it and do something that really shows you as a human so I'll warn you with that that you can do it to write everything but I prefer to do it to outline as you can see here provide me suggestions and those kind of things I don't know if I'll spend time pulling up these links but um I'll just talk about them at a high level and you can feel free to go to these links check them out I just made them like publicly viewable chats that I've had with chat gbt in the second one and then uh Bard which is Google's uh large languages model assistant in the first one and we can come back to those later on if you'd like to see them as well but uh either way the first one I use it for a job searching it can actually be very good and specifically barred as of today which again these things change every 25 seconds so you know don't quote me on this but as of today Bard is the only one that is connected to the internet so it can actually search for you oh Bing chat can which is built off chat gbt but Bard is another one that can uh search the internet for you and so it can be really good at just saying well uh so what I did in this one and maybe I will actually pull it up if I can get my screen shared here pretty quick let's see let's back out of the this and then we'll pull there we go so you should be able to see this now so here I gave an example or I gave it a prompt and I said you know imagine I'm currently a sophomore in college at Utah State University studying marketing with an interest in aviation uh and trying to find internships that overlap with my experience and so you know I can also and I probably could have done here if I had a resume that reflected this uploaded my resume as well to really allow it to you know pull out the details the keywords the techniques the Technologies all these things that I know that maybe could match in these but as you can see here even with this generic description it pulled out quite a few different uh internships that are currently available um in the Utah area you know Hogan Provo Salt Lake City at a couple different airlines as well as some internships and opportunities here at Utah State in the aviation management Aviation safety space so some really cool opportunities and then I dove in deeper and I said well actually I'm looking for more of a startup environment where I can have an opportunity to grow and then it pulled that out as well and so you can see here I mean I could have always gone on to LinkedIn or you know any other job search board and found these same internships but I have to you know create filters and it doesn't really capture maybe my unique perspective and the way I want to search and so I have found personally that using these tools in this way I can actually speak to it as I would speak to a person right I can say the things that matter to me and the things I want to look for maybe pull out the fact that I want to you know not just work in the aviation space I want to specifically work and you have the autonomous drone space it's harder to search for that in you know filters and keywords much easier to use in natural language and so this is a great way to leverage these tools to quickly and efficiently search through it and the best part is I can rerun this as many times as I want so tomorrow I could come in and run the same exact thing and it probably generates some more and for that matter these are stochastic so that means it has some Randomness in there if I rerun this exact same prompt it might give me a different list each time so there is that aspect of it's not comprehensive necessarily but it's a great starting place it's a great place to start exploring and researching and just getting a lay of the land really more than anything uh this other one that I wanted to pull up here let's see this one is one that I really enjoy doing with uh Chad gbt and this is something that I say will not only apply to your uh job search but also just in general to your you know educational Journey we're in an era now where you know I mean my job again is at risk in some sense of you know we have to reflect on what it means to learn and this is a perfect example of this um the era of just saying you know here's a textbook read it and you know come to class every day I'm going to lecture on this topic I mean that's kind of gone we really need to focus on individualizing our learning um targeting the things that we understand and finding the gaps in our understanding and building on that ourselves and these tools are so great at helping you do that so this is and I'll kind of zoom in a little bit this was a specific example that I had um so I work with the Aspire NSF engineering Research Center here on campus and we do a lot of electric vehicle research and there's been discussion recently on you know hydrogen fuel cells versus Dynamic Wireless power transfers so charging coils in the ground as we drive over then which one is a better approach for um heavy duty truckings like 18 wheelers and I realized I didn't really have a super deep understanding of hydrogen fuel cells and so I said okay well I have some time this summer give me a six week study plan that goes from very little understanding of hydrogen fuel cells to full-scale fuel cells in practice right and you can see here that carved out you know this kind of week by week outline relatively generic but it starts to get in some specific details here around week three um and broke it down again by each day of the week and so you can see it call that a significant amount of study each day I could have probably asked it to reduce it to two to three days and it would have done that as well and here I said create readings study and assignment list for week one and at this point it started to pull out textbooks which were I think both of these are real textbooks that it mentions in this uh information here as well as reading plans study plans assignments and so you can start to see that this is exactly what you know the job of a professor kind of used to be and again if we talk about job opportunities I have to now reflect on and say well if it can do this what what is my role in this where do I fit in this and you know there is still this value of human interaction and more than anything kind of the um guidance and scoping of your educational plan and also the ability to interact uh with somebody who has had the human experiences of this so there are still aspects of this we have to Define as professors but more importantly is to not you know say well guys don't use this to to learn stuff I actually would love to see this right I would love for you to say great I don't understand anything about this I maybe feel a little bit out of my depth on this topic and so now I'm going to use this to upskill myself and that's really if we go back to this idea of what does it mean for um job search and I'll kind of go back to the slides here what does it mean for my job search and career advancement well now you have the best tool in probably history to learn a new skill instantaneously right you can say well great I've been working in marketing now or I've been you know taking all my marketing classes but I sure wish I could have done some machine learning stuff because I think that's going to matter for what I want to do or maybe I want to learn more about Google analytics or whatever it might be or rather than having to search for an online Coursera course and maybe find the course here at the University that you know close this aligns to it but you maybe don't have the prerequisite for it you can just Define your outline here right or maybe figure out the skills necessary for the prerequisites that you don't have for the course you want to take build those now and get to that point or even imagine you're on the job and this is one of the ways that I have found you know the best way to advance your career is getting into a company company that has some sort of upward Mobility and then recognizing what skills do I need to build to get to that next step usually within a company they're not going to be as interested in you may be saying well I want to move into this field or I want to move up to this space and so I'm going to go back to school for the next two years to get to that job they want to see or if you want to do it they just kind of need to see that you are capable of doing that task especially if you've already formed that Rapport within your company and so now you could say well I'm in an internship working in this space I'd like to work in this department in the company or this Division and so what are the skills I need to get from here to here maybe even put the job descriptions right the job description for the internship to as one example the job description for your dream job and the other example and say what if the six week or the six month you know study plan that I need to do to get to that point you have those tools available available to you today again this is not something we used to have or you know had anything close to this and so really start to leverage this both in the academic sense in the school sense but also in the job sense the ability to teach yourself is gonna you know really start to separate uh individuals in the workplace and so that kind of brings me to this point here I think it's my last slide second to last slide it's my last slide here um just thinking of again how to Future proof your career um the first point there probably being the most critical is recognizing that the nature of automation is to remove tasks that are uh you know easily defined and then you know kind of or better something that we can develop Technologies around to Define and then iterate and Advance right the soft skills right the things that we've historically called soft skills mostly because we can't Define them those are going to be the human aspects right because we're not good at defining what it means to be creative or what it means to be Innovative or to you know understand and communicate and collaborate with your teammates because we can't really Define that concretely it's not likely to be automated anytime soon these tools are all I think these AI tools are all kind of based around this idea of optimization and so without the ability to optimize towards a Target they're not going to be able to solve it and so this is going to be the part that really matters for us as humans and this doesn't mean that maybe you don't need technical skills it just means that you need to understand your domain in the context of these skills right understand how to consolidate ideas from different regions and different in different uh domains into a single thing and use your skills and your AI assistant now which is that second bullet point to build these things faster and to really do um leverage these as assistants use them to automate your daily tasks use them to you know help you at home even but just start to get very familiar with these ideas and these tools in their ability to really just make everything more efficient but I will kind of Hazard one thing you can notice there under that second bullet point over this under the soft skills uh there is this aspect of bias that's still baked into these AI models this is not intentionally done but it is a nature of the way they're trained they're trained on you know in the context of a thing like GPT it's trained on as many documents as they can really collect from the internet and as you all know there are opinions on the internet because they're humans on the internet and so therefore there are a lot of opinions and biases that are kind of inherently baked into these models which means that they will have some sort of impact on their decisions or on their outputs a good example of this is in the HR space there's a lot of tools that have tried to be developed to automatically identify likely candidates from resumes these have pretty famously failed many times over because they're going to only be trained on historical data well if there was any sort of bias in historical hiring practices which by construction they always are because we are humans and we all have biases those will be reflected in the outputs of these models right of these machine learning of these AI tools and so because of that you're going to miss out on great people and great opportunities to hire interesting people and without humans being in the loop on this process and understanding that that is a bias that will exist in these models we won't be able to kind of move out or move outside of that or improve upon that or iterate on that and so that's again one of those soft skills is understanding the limitations of these tools in the context of your domain and recognizing how that might impact those Downstream decisions and how you can you know still leverage them as a tool but be part of that decision making process okay so there's kind of that that two-sided uh of this idea and the last bullet point that really is an impossible task to understand all the technical advancements in the space but do your absolute best to understand you know why these tools matter which tools matter and what it is about them that is even a development as I said in that last you know slide I showed you three different large language models barred Claude and Chachi BT Again by the time this goes live that could be probably 1700 more and the question of which is best which tool you should use is an impossible thing to answer I can't Point towards a single tool to use to help you in your career Journey you know whether it's finding a job or you know moving up in your career but I can at least say that well an understanding of these tools and where their limitations are where their opportunities are and what they can do that'll at least give me enough information to understand when something new comes out that actually matters right that's not just a sales product I mean that's another aspect of this you will have to start to deal with it's already happening now is that a lot of people are building tools that are maybe not you know something meaningful at least I'll say to be nice and so um you know we have to understand where we're actually getting value from it where we should be putting our dollars towards what will have some sort of return on investment and so an understanding of these technologies will be critical in that and that's where again I'll say you know please come take our data analytics courses we have quite a few specifically I talk about these tools and techniques of large language models in data 5610 which is our deep learning course deep learning models are what underpin these and we also have a text analysis course that's taught by my wife uh Carly Fox and that's an outstanding opportunity to really get a sense of you know how do these large language models even work right how do we understand text in numbers right like how do we form any or pull out any sort of patterns from text because we have to turn it to numbers so all that stuff will give you at least a high level of overview of what's possible and given today's Technologies and maybe the future Technologies and then why that might matter uh in the coming years so really my main takeaway is soft skills really you know hammer down on those work with people understand what makes your unique skill set matter right what is your unique skill set um and leveraging an AI whatever your choice of tool is as an assistant throughout this process right really see the way to make yourself more efficient in this world so I'll kind of leave it at that open up to any questions from you Valerie um but hopefully it gives you at least a starting point to work from I think I can't hear you quite yet Valerie I don't know if you're you did perfect thank you you think after covet I would have that down pat yeah but thank you so much for taking the time to share with us this exciting topic of AI and how we can use it for career Readiness and some of the things that really stood out to me that you said is really focusing taking the time to focus on those soft skills because while a lot of the repetitive things may be replaced by AI there will always be the need for that human ability 100 right and so thinking kind of then what opportunities lie ahead is so exciting exactly it's one of my favorite things about it I think we again Define these again as soft skills because we didn't know the words uh to them the same thing happens with the soft Sciences but turns out those are the ones that are harder to do uh the ones that are harder to automate away so you're definitely in the right place if you're thinking along those lines we we as data scientists did a great job of automating our way our own jobs uh so now we have to reflect on that still right exactly and I was thinking about that too when we meet with students I've actually been using uh forms of AI to help students with their resumes and cover letters as you suggested and I was thinking with the amount of time that we save doing resumes then we could as you were saying you know divert our time to be productive in other ways we can spend more time networking because networking is so critical oh 100 and kind of along those lines one of our questions is if you're familiar with any AI tools to facilitate networking yeah so recommendations we've all been using the best AI Tool uh for networking and that's LinkedIn uh it's the funny thing about it but that is really their product is how do I connect with the right people at the right time and so you know there hasn't been anything close to that today mostly because what we're talking about here is not necessarily a technology so much as a data set where can I quickly and efficiently find all the people that matter for this process um you know there is this aspect of you know helping me you know come up with ideas or ways to connect with them or you know send or paragraphs to send to them that'll somehow convince them do it but you know there is no Silver Bullet in that sense but the really best thing you can do is represent yourself well on LinkedIn um the algorithm sure does love interaction uh you know a good example of this is we had a team do really well in a competition here and you know I was tagged in the post about them as I was mentoring the team and the amount of interaction on my page after that was Mindless far and above anything I'd ever seen now unfortunately I wasn't looking for a job at the time because that would have been the perfect time right to you know have all the views and all the eyes on this but that is just the nature of these platforms they love interaction and more than that is these you know first and secondary degree connections that we have uh those can be really great ways to find where could we fit in not only from a you know ideal job standpoint but where do I have some connection that can maybe help me understand if I fit in well there if I'd like it there and also if you know they'd be interested in hiring somebody like me so if I see that and this has happened to me is you know if I'm looking at jobs if I see that I have a first degree connection there that's going to very much change my consideration implying and also will heavily impact my application itself and for that matter my last two jobs prior to the university were both sourced through Linkedin one was an application I put in uh for an internship and they saw that I was a PhD student and that I had experience in the space and so while they weren't looking for a p student they were able to open up another position specifically for me and the one after that was through a recruiting agency that was looking for people with my set of skills in this space and they saw all that on my profile very clearly laid out and they saw that I was working in a government contractor before so I had a security clearance and so there's a very clear way for them to reach out to me contact me and connect me with the company uh which led to the job so you know the best tool for networking the best AI tool for networking is the tool that has the best data set and so that's how kind of like future proof that answer is to say that right now that is LinkedIn that might change in the near future um and the format will likely change too in how we interact and we're starting to see this already as well too you know the resume and cover letter important they're great ways to you know reach out a company that you have no other way to contact you or contact with but companies don't really love to look through all of those all day every day and they tend to themselves also scroll LinkedIn with just being relevant on there having your presence there known can be super critical connecting with people and also connecting people with people meaningfully not just scatter shot connecting to nobody loves that either I will very quickly ignore somebody I don't know um so again even if you're a student and you're interested in connecting with me or any other Professor if you contact us or connect with us and then also put in a little blurb of like what you're looking for from me because you might say hey I'm very much looking to see if you have any connections in this space right in the aviation industry as we mentioned before and then I can say actually great yeah I'd love to meet with you and I have some friends that work at Boeing and so you know let's see if we can set up some sort of interaction with that those are way easier connections to make and ways your jobs to get than the you know application out to everybody and I've done that path I've probably applied to all the order of thousands of jobs that way but very very few of them have laid at me jobs yeah that is fantastic I love that every everything would want all the students to know use LinkedIn and yeah make those connections meaningful fantastic um perhaps in the interest of time we'll [Music] um maybe ask just one other question um and it's what would be any ethical considerations or concerns related to using AI yeah this is a this is a really good question because I've seen because I tried this actually I tried saying you know this exact same question I posted before of here's my resume here's the job description help me make changes to make myself more relevant if I'm not very careful in how I phrase that it will very much fabricate my resume um and you know maybe a little bit of fudge in here there is I can you know support it enough but at some point if it suggests a technology I don't know there's nothing I'm going to do to be able to back that up and you know having been on the other side of hiring uh you know junior scientists and that kind of stuff you very quickly pick up what they do and don't know what was a little bit fabricated on their resume and that never uh never looks good um so that you know that is the real risk to me you know from an ethical standpoint you have this other idea of you know you don't want to be using the Earth as I said before these tools can and will reflect biases that have exists in the data set if you are using this to create your resume or cover letter it will say what is most common right that is kind of the nature of a bias right whatever is you know the prevalent idea whatever is the default idea it will say that and so that's where that fabrication really comes from um and and what you want to be careful with uh in the producing of this so I would say again really really try to think about it as an assistant right think about it like you have an intern sitting right next to you that's helping you with this process how would you phrase it to them and that is how you can really start to like steer yourself away uh from doing that um you know there there are plenty of other things they tend to the the real ethical considerations to come from the other side of this which many students might find themselves on very quickly of in the actual hiring process uh in the writing of job descriptions in the uh choice of which resumes to review and what matters in a resume that's where a lot of those ethical considerations come in from a like social standpoint but really for you as an individual applying to jobs just try your best to keep things as in your voice as possible you don't want to especially in a cover letter you don't want to write something that all of a sudden sounds like a New York Times article and maybe it's written beautifully but it doesn't really reflect who you are how you present yourself and so that's going to come through that sort of dissonance between what you put down on paper and what you say in person really does not come off well in an interview and so do your best to get your foot in the door for sure but represent yourself in the best way possible and if you're lacking in an area that's a great opportunity to use these tools to try to build yourself up in that area such that you can be more honest in that in that representation excellent yeah great advice and just one more quick question is there a time where you would say that it's worth paying for a service yeah this is uh the part that's changing fast um you know every couple weeks there's usually a major update um some new features some do beta features that come in chat gbt that usually are only reserved for um paying customers and it's one of the things I don't love about that platform as well too because this is an issue that we're having to face in the data science if you talk about ethical considerations um the opportunity or the ability to be able to pay for these tools which is not an insignificant amount right at 20 bucks a month the ability to pay for these tools should not preclude you from the opportunity to use them so there are thankfully a lot of efforts to democratize these tools but there are there can be some advantages I'll say to paying for Chad gbt specifically as an example right now I think the plugins and things like the code interpreter are only in the paying version those can really help you in certain positions in certain things you're doing but I I would say that for the most part if you are in the job search process you're probably fine without paying for chat GT especially because again as of today August 8th I can only say that the it is not connected to the internet it was for a little bit many roofed connection to the Internet so they can't search the internet for you whereas Bard can and again as of today Bart is free and does Bing chat is free and does and so that's really what's going to matter more in the job search process is being able to search internet and also being being able to upload files right being able to provide your resume uh and those kind of details in there are going to be critical uh the other major thing that will likely be part of payment plans in the near future and already kind of is with chat gbt and this is a little bit more subtle and under the hood but is the token limit you can kind of think of it as a word limit and that is you know again how much how many words can be not only in the input but also the output so the sum of those is the token limit and so if you have something like I think gbd 3.5 defaults now to a 16 000 token version maybe um that means that across my resume my question my job description and the answer I can only have 16 000 Words which is quite a bit but other things might allow you to have hundred thousand words so like quad for example has been exploring 100 000 token model which means I can maybe upload you know five different job descriptions and consolidate them down to one or you know write a cover letter that is much longer you know take my whole dissertation potentially and condense it down into a one pair graph blurb that you know better describes it to you know a specific job so there are some features that may be advantageous for you and this is where again it really comes down to understand the tools and their capabilities and what you need out of it but if you're just putting a generic question in there most of them off the shelf can give you a decent answer it's really these more kind of edge cases and these more special prompts that you're putting in there where you might need some more capabilities out of them perfect once again thank you so much for taking time to share your time with us and especially your expertise we're so fortunate to have you with our team here at USU and just a few announcements uh please join us for our career Expo it'll be Friday September 22nd from 11 to 2. it'll be in person and so there won't be a virtual component this year and to prepare for that if you want to stop by the start desk between 11 or sorry between 9 A.M and 4 P.M there will be

either a peer mentor or a career coach to help you with your resume so that you can prepare to meet with those employers and then lastly is just to check the Huntsman calendar for other upcoming events and hope you have a wonderful semester thanks

2023-09-15 08:53

Other news