0:00

[00:00:00] [00:01:00] Welcome back to another episode of Interface. I'm your host, Jennifer Plummer, and I'm joined with Sierra and Maddie. And today our guest is Bobby Kyle. He is an IT advisor at Central Louisiana Electric Company. He has over 25 years experience in engineering and information. Information technology leadership roles. jenni_1_01-26-2024_121542: He currently works as an advisor, assisting leadership with implementing IT strategies with business goals. In addition to this, he serves as a subject matter expert in the areas of quality engineering and works with cross-functional leadership to deliver high quality software optimized processes and improve time to value. Previously, he was the global head of quality [00:02:00] engineering at pros, where he was an active member of Empower who also sponsors this. Podcast, um, the Employee Race Resource Group for black employees. He is passionate about people development and is a champion for diversity, equity, and inclusion. Bobby graduated from Prairie View a and m University where he earned a BSS degree in mechanical engineering, born and raised in Houston, and is a proud husband and father in his free time. Bobby loves to cook, spend time with family, and is an avid car and motorcycle enthusiast. In addition to all that, uh, interface has been doing a networking challenge. We asked some of our executives to nominate wonderful guests, and was nominated by Sue Neal, our Chief Product Officer. So he is the first of, um, and there will be more to come. Welcome to Interface, Bobby. How you doing? squadcaster-dj4c_1_01-26-2024_121540: I am doing great. How are you? How are you doing? It's great to see you all [00:03:00] again. It's been, uh, it's been, it's been a while, but fantastic. jenni_1_01-26-2024_121542: awesome. We're doing great. We're happy to see you. Where we'd like to start is, can you share with us your, uh, how you came to be interested in technology and have a career in technology? What was your journey? squadcaster-dj4c_1_01-26-2024_121540: Good question. So, I graduated from Prairie View, uh, in 1994 with an engineering degree, a mechanical engineering degree. And the way I get into technology is coming outta school. The very first job that was offered to me was, uh, a job as a flight verification engineer. at nasa. So, um, I started off verifying Flight Path software, um, in 1994, the rest is history. I never looked back. Um, since then I've had, you know, roles at, uh, engineering and, um, it, related roles for the, 30 years. So [00:04:00] that's how I got into it. Uh, when, when I, uh, first started at nasa, it was interesting. so I decided to stick with it. That's, that's, that's how I got into jenni_1_01-26-2024_121542: How did you gravitate towards the quality engineering side? squadcaster-dj4c_1_01-26-2024_121540: So as a flight verification analyst, what that is is you're, this is back in the nineties now, so you're, you're getting data dumps and you're analyzing that, that flight software, you know, the trajectories, the abort pass and things like that. So that actually started off as a true. Uh, very low level, um, quality engineering job. And so my career from there jumped into, uh, the healthcare industry and then, um, you know, uh, system software where I got into management. So that's, that's how I got how I grab gravitated to, uh, the quality engineering field. mattie-cakes_1_01-26-2024_121542: Yeah. And Bobby, for people who don't know what quality engineering is, I, so I work with QE at. As a pm [00:05:00] um, can you maybe tell us what quality engineering is and how it differs across those various industries that you were in? So healthcare to software. squadcaster-dj4c_1_01-26-2024_121540: All right. Sure. Great question. So let's talk about quality engineering versus quality assurance, right? Uh, they're different. And so starting with quality assurance, quality assurance is basically taking, uh, business specifications, functional specifications and validating functionality, um, uh, either functional non-functional, which is, um, you know, manual automation or performance and, and load testing versus the requirements. You look at quality engineering, it really spans outside of what you guys typically think of, uh, qa. It starts at the very, very beginning shift, left. and we work on identifying processes throughout the entire software development lifecycle that will improve the quality of, uh, the software or the deliverable. So that's the difference [00:06:00] between QE and qa. Now, when we talk about the difference between, um, the different, uh, excuse me, QE across the different industries, it's really about the software that you're testing, right? Um, if you look at the roles in it, uh, those are companies where you consume software, you integrate them, and you test them. when you look at the engineering companies that I've worked for, uh, these, the software's built in-house, so it's a little bit more complex, meaning that, you are actually part of the, the engineering under or development lifecycle. Um, and so those are the differences between, uh, the various industries that, that I, I've, uh, I've worked for. jenni_1_01-26-2024_121542: you see a, because you mentioned you, you know, being at nasa, which is aerospace, and then, um, healthcare. Is there any. Fundamental differences between those two industries or is being in the role qua quality, um, engineering kind of the same. squadcaster-dj4c_1_01-26-2024_121540: It is the same, Jennifer. It's, it's really the. [00:07:00] NASA was a little bit, was way more complex than, than healthcare, of course. Uh, but your, your role a day in the life of what you do, uh, is typically the same. It's just what are you actually, uh, testing and, and validating, uh, the software solution or solution is what's different. jenni_1_01-26-2024_121542: So what would you say is your daily, um, you know, day in the life, you know, what, what does your typical day look like as an IT advisor or head of quality engineering? I. squadcaster-dj4c_1_01-26-2024_121540: All right, let's, let's, let's start with, uh, well, let's take head of quality engineering. the very first thing that I do, first thing in the morning is I check my email, uh, just to ensure that nothing urgent came in overnight, after hours, or what have you. Um, I also looked through my email to make sure that, uh, if there was any emails that I did not respond to, uh, or any questions that needed answering. Um, that I respond to those first. Um, the second thing [00:08:00] that I do is I check in with my team. So I check in with the, the QE leadership team, uh, especially, uh, the ones that are overseas, perhaps, uh, to make sure that we don't have any hot burning topics, any challenges, anything, uh, that needs my, my urgent attention. Um. The next thing throughout the day. Of course, as you know, Jennifer, 'cause you are in these meetings as well. Uh, there's cross team collaboration. So we talk about, you know, uh, projects that are, that are, um, currently in progress and we talk about pro uh, projects that are coming down the pipe. we make sure that we have, you know, resources aligned. We understand, uh, the requirements and we have those types of discussions. Talk about the risk also talk about, you know, what is it that we need to consider. at the very beginning in, in terms of improving quality, whether that's extended time for automation, um, bringing performance testing in sooner rather than later, uh, or just making sure we have enough people, meaning sometimes we have to [00:09:00] slow down projects. Uh, to speed other ones up. So it's, it's more of that. And then, uh, the last thing that I spend time on, uh, throughout the day is strategic planning. We talked about a little bit of that in terms of, you know, what, what, what will we be working on in Q1, Q2, Q3. Um, what is the strategy for the QA organization? Um, as you know, we, you know, for example, we went from mostly manual testers to automators, uh, and right before I, I left, we were looking at ai. So, um, that's a day in the life, a typical day in the life of, of, um, a quality engineering, uh, director. Uh, and of course it's Track 1: Oh, can we double click into the strategic side of QEQA? Um, from my understanding, you know, you have a. Like you said, the functional idea of what should happen, and then you, uh, you have QE that go in and kind of test is, is things breaking or the behaviors as [00:10:00] expected. So that's kind of maybe the surface level that I. I think the majority of people understand, so the automation, the, the AI aspect, like the actual strategy of queuing and not just, you know, we're just gonna throw people to make sure things aren't broken. Can, can you talk about how you look at the strategy side of q uh, quality engineering? I. squadcaster-dj4c_1_01-26-2024_121540: Right. So, um, you know, every company that I've ever worked for, cost has always been a factor. So, uh, the old ideas of just throw more QE at a problem. Uh, that doesn't work today. Um, so you have to think of ways that you can do more with less, uh, and to do that, uh, what we, what, uh, one of, one of the ways that we do that is we bring in automation. And automation, um, brings down, uh, the cost of your regression, uh, test cycles. And so that is, that was the focus, is to go out and [00:11:00] automate as much as possible so that we can run that. It's, it's, it takes less time. Uh, it's lower cost. it's, and it's, um, it's more accurate. So was the first thing that, that we really wanted to do is bring down the time to value, uh, from a testing perspective, uh, bring down the cost of testing. Um, and it's, it's significant. It could be, you know, as large as 80% time savings in some, in some instances. Um, and then as you continue to go down that path transformation, um, you need to look at better tools and the tools that are available today. Uh, most of them are leveraging, um, AI and AI is important. Um, there is no limit to to to what you can do with ai. Um, and one of the things that, that we're looking at is using tools that, um, there's a Jira plugin that you can use that will read a user story. And then create manual test cases. Now those manual test cases need to be [00:12:00] verified, of course, but using tools like that to generate the manual test cases. And then you can use or leverage ai, uh, to write automated scripts, you know, especially if you have, uh, quality engineers that are, um. That are less technical. Um, and then last but not least, using things like machine learning, uh, to read, uh, data, you know, that's coming into your support organizations to better understand, uh, where your defects are coming in and, um, report out on your metrics. Track 1: Yeah, that was great. But I'm gonna hit you with another one because you hit, you hit me with all this terminology and vocab and we're trying to demystify technology. So can you maybe give me a very quick. 1 0 1 on QE of these, the vocab regression, functional smoke testing, all that stuff. Automation, what? What exactly is that? Does that mean. squadcaster-dj4c_1_01-26-2024_121540: All right, so let's start with, [00:13:00] uh, the first, the first, uh, thing you asked about functional testing. testing is basically taking, uh, a business requirements document. Reading it and then writing a manual test ca test case to validate that piece of functionality or that workflow. Uh, that's typically, those are new test cases. we talk about regression test cases, those are test cases that you've already executed at least once. So those are regression test cases to make sure that as Jennifer and her team. As we're adding functionality, we did not break existing functionality. So, uh, it's very important. I mean, and Track 1: Especially for Jenny's team, right? Is that what you're saying? jenni_1_01-26-2024_121542: Yeah. Why? You gotta put us on the spot like that? Huh? What? squadcaster-dj4c_1_01-26-2024_121540: No, she's, Jennifer's always been great, but, but no. Um, and then when you talk about the different levels of testing, when you talk about full regression, what that means is every test case that you have [00:14:00] in, in your test suite, when you look at high risk regression, what that means is, you know, you've worked with your developers or your architects to say, Hey, here are the things that, that are at, uh, the highest risk, and this is what we wanna focus on. Uh, some of these other test cases are low risk because we haven't touched that area of the code or touched that integration or what have you. So, so think of high risk as a mid, mid grade, uh, test suite. Typically 60% of your test cases or, or so. And then a smoke test is when we get down to like the last few defects, uh, in a release and they're very targeted and you basically run a sanity check plus validate. Uh, those last few defects to make sure, um, that nothing was broken. So it's the lowest level in terms or the lightest level of, of test coverage that you're gonna perform. jenni_1_01-26-2024_121542: What advice would you give, um, up and coming, uh, who [00:15:00] are interested in qual eng engineering? Is there some sort of career path they should do or things that, specific things they should study? squadcaster-dj4c_1_01-26-2024_121540: I guess, let me answer a different way. What I would say to anybody who's up and coming and wanting to get in into leadership Number one, uh, make sure you get some coaches. Make sure that you, you reach out and you find mentors early on in your career that can help guide you, down, down a, a leadership path. Um, the other thing I would say is before you pick quality engineering, uh, development, support or anything, I think you should work in, in, in different organizations so that you understand not only the relationship, but. You don't know what you like until you've tried it. So once you make a cycle around, um, you know, the different cross-functional groups, maybe you may, maybe quality engineering is what you, you want to, um, to land on. the next thing I would say is [00:16:00] make sure that you stay curious and make sure that you stay abreast of, uh, emerging technologies. Um, that's what I would tell anybody that wanted to come through this path. siara_1_01-26-2024_121542: Speaking of emerging technologies, I, I was interested, you touched a little bit on AI as you were talking through, um, sort of your day in the life and I'm interested, obviously ai. AI is impacting us all and how we work. Um, I'm interested in how you and your team may be leveraging AI to be more effective, um, in your role and roles. squadcaster-dj4c_1_01-26-2024_121540: in my, yeah, in my current, um we just, at the very cusp of of, of getting into. Um, to ai. Um, if you look at it more broadly again, um, I think that that, you know, most of the, the, the, the places that I, I I've worked for or talked to, I think they're trying to figure out what is the right path, you know, in terms of [00:17:00] risk, uh, and, and leveraging ai. I mean, I think it's, there is unlimited untapped potential. Um, but, you know, how do you get into it? Um. The right way, I think is the, the, the, the biggest challenge. Um, I know when I, when I was at pros, we were looking at it again, um, to analyze support data, uh, to see if we could use that to, to, to create defect leakage. Uh, also for some of our, um, like I said earlier, some of our less, less technical resources, we, we were using that as a training assistant to say, Hey. You know, write me a price method or, or how do I write, you know, a, a end, end workflow for a price method? And you could use that and, and, you know, any of these AI tools and it'll spit out code snippets, that can help expedite your testing. siara_1_01-26-2024_121542: Awesome. I'm still also trying to find small ways that I can leverage ai, in my day-to-Day work. So yeah. I was just interested [00:18:00] in if you guys were, I. Starting to use that to become more effective in your day to day. Thanks. squadcaster-dj4c_1_01-26-2024_121540: Yeah. So Sierra, just to, to make sure that, that I, I close on. You know, I use it every day for me personally, you know, and, and, and to help with my career. But one of the things that I would advise is make sure you validate to the best of your ability, the, the, the information that you're receiving from, from tools like chat, GPT or bar or what have you. Because I have seen mistakes and that, that's what scares me, is when you, uh, come back and the, when the information comes back and it's incorrect, it it Track 1: I love having the conversation about like personal use cases for, you know, gen ai, H-I-G-B-T. What are your use cases? You say you use it every day. I also use it every day, so I'd be curious to hear more about that. squadcaster-dj4c_1_01-26-2024_121540: So I use it as a reference, you know, I mean, one of the biggest things that, um. That I use, like I said, I like that I use it for just a [00:19:00] reference instead of, you know, typing in, um, um, you know, in a, in a browser, you know, a question, you can just type it there. Uh, I also use it for things like, uh, I do a lot of work in, in Excel and sometimes, uh, I may need, I may need help with, uh, an Excel equation or something like that. So I use it that way. Um, I use it just as a, for me, in my role at, at my, at my level, I use it for general information, so. I, I don't do a lot of technical stuff anymore. Track 1: I think there's a future where, you know, everybody's leveraging AI to do everything they want. So you, you know, you have specific AI for specific use case. You know, the one that you used is, is the, you reading user stories, populating, uh, test cases. Probably not gonna be good to help you find, you know, a pizza restaurant or something. Or order a pizza. Um, so yeah, various different use cases. I, I, I see a future where, you know, everybody's using gen AI and then everybody else becomes consultants [00:20:00] to that single person. So I'm, I have Gen AI to get me 80% of all these areas that I need, and then I have. The experts to help me finish that 20%. You know, there's ways to do this better. Maybe change this wording, it should function like this, but you are able to make or allow people to really just focus on that, that you know, that 80 20 rule, that 20% of what they do really well and what they do valuable, and then a single person can get everybody to that 80% and then everybody can, the rest of the people can focus on that, that 20%, the actual value. Do, do you, what, what are your, what are your thoughts on that in terms of, you know, from a, a QE perspective? I think, you know, there's, there's two narratives, right? Uh, and maybe you can talk about it to the automation side where you have like j AI and it's taking everyone's job and no one, no one's gonna have a job anymore, versus j AI is just helping everybody do things quicker and better. And you [00:21:00] shouldn't actually, just because you can 10 x one person doesn't mean you should get rid of nine people. That just means you can 10 x 10 people and you have a hundred percent more output. Can maybe have you faced kind of something like that in the, the realm of queuing when you move towards auto automation. squadcaster-dj4c_1_01-26-2024_121540: So are you asking about Track 1: I I, think I'm asking about AI specifically, but then using maybe your past experience through, through going from, you know, manual testing to more automation, how there might, you can do both in parallel and you don't actually need to worry about some of the, the, the, the AI fears. I dunno if that works. If you can, if you can put that together. squadcaster-dj4c_1_01-26-2024_121540: All right, let's, well, lemme, lemme let, let's break it, break it into to, to two parts. So, you know, when I think of, of qe, um, pre onset of ai. The way that I like to structure my teams is that you have functional [00:22:00] sme, functional SMEEs, and those SMS ideally should be as, as, as good and as deep in an application or solution as a BA. I mean, they need to really mirror a BA in terms of their functional knowledge of how an application works separate. And apart from that, you have automation experts. These are people that, you know, go very deep in, in automation and coding. to say that is what I'm trying to say is that there is going to always be a need for manual and as well as automation, you are always gonna need somebody to write the test cases before you automate them and to, to make sure they're correct. So, um, I think, I think there's, there's enough space for, for both to live in, in this world of quality engineering. Um, but you have to be one or the other, in my opinion. Um, bringing on ai, that's only going to help. Uh, you're still going to [00:23:00] need manual people to validate the information that you, you, you're getting back from ai. You're still gonna need people to validate. Um, the snippets of code or the, the, the, the, the programs that AI is bringing back to you. Um, but the world is changing. I mean, you're going to have to, and, and we're gonna have to change as testers, meaning that you're gonna have to make sure that you bucket yourself in one of these positions, um, because AI is coming and jenni_1_01-26-2024_121542: Bobby, what programs do you think companies should implement to increase black talent in technology? squadcaster-dj4c_1_01-26-2024_121540: Oh, that's a good one. Um. That's a good one. So I think, you know, the focus on, on black talent, uh, I think that needs to be part of a broader DEI, um, initiative or program. I think programs to increase the number of blacks in technology, I think it needs to be, it needs to start at the top. It needs to be [00:24:00] supported, um, from the very top executive leadership. Um, and it needs to be very intentional. I think that program also needs to include what is the target, and that's all levels. That's not just number. How, what, what should be the target that we should see in, in management? What should be the target? All levels of management and, and, and then our staff, right? Um. I think the, that program should include Jennifer, more comprehensive training for, uh, not only, um, for everyone, so they understand the importance of de de and i, de and I and how, you know, how it benefits the company. And, and that needs to be, like I said, it needs to be very comprehensive training, not, not high level light training. I think that one of the things that, that would also help, you know, uh, more represent representation in tech is having, uh, mentors and coaches, [00:25:00] uh, for, for, for blacks. You know, uh, I think that if you have some money that you can talk to, you can understand how to navigate some of the obstacles, uh, that we often see in technologies as it as it relates to blacks. I think other things that you can do is, uh. said it starts with leadership, Jennifer, but it's also on us. Meaning we need to use our networks, to identify people that we think would be good fits within the company and promote that. You know, I think, uh, the program also has to have, uh, black, uh, representation throughout the entire hiring process, meaning that as part of the panel, you need to have representation. If not. Um, I don't see it's gonna be, you know, I don't see if, if, if I don't see, uh, a way that this is going to, um, I think it's going to be the sa the same way that we've seen before, meaning that, uh, for whatever [00:26:00] reasons, you know, we're not selected. I think that we also need to expand your partnerships within, uh, HBCUs, uh, their tech programs and their, their placement departments. Make sure that, um. You have solid partnerships and, and, and that we're sourcing from the right areas. and then last but not least, I think that you need to make sure that, uh, we have a way to measure that we are improving. So I think that's what a good program would look, um, to me. jenni_1_01-26-2024_121542: It's, yeah, it's, it's very comprehensive and yeah, you kind of touched upon the levels, right? Individual contributor to manager, to executive level support from the top, but also, um, you know, reaching out to each other, making sure you have mentors and, and coaches and measuring what matters. So you had a very, very comprehensive answer. squadcaster-dj4c_1_01-26-2024_121540: [00:27:00] Well, you know, this, you, we, we've worked jenni_1_01-26-2024_121542: Yeah, squadcaster-dj4c_1_01-26-2024_121540: so, you know, it's jenni_1_01-26-2024_121542: we've had these talks. Yeah. squadcaster-dj4c_1_01-26-2024_121540: know? Um, and so, yeah, and, and even extending beyond that, I mean, there's, there are programs, there's STEM programs in high schools and, and inroads and, you know, if, if, if this is an intentional, um, initiative. We need to plug into those as well. I mean, there's lots of ways to find black talent. Um, you know, if you're, if you're really looking for. jenni_1_01-26-2024_121542: A hundred percent agree. siara_1_01-26-2024_121542: All right, so we are now transitioning into the bam, bam, bam, the heat check portion the, of today's episode. Um, the heat check is a segment where the interface co-hosts, uh, share an interesting topic. technology or uh, black culture. Bobby, as our honored guest today, feel free to chime in on any of the topics that are brought [00:28:00] forward today, and I'm gonna go first today. Um, so I was browsing this morning, um, for my heat check topic, and I found an article, um, uh, on USA today about the Apple Macintosh is turning 40 this year. Who, who else feels old? Who? jenni_1_01-26-2024_121542: I siara_1_01-26-2024_121542: Who feels old here? jenni_1_01-26-2024_121542: That's, Track 1: an Apple Macintosh? jenni_1_01-26-2024_121542: a given. I wake up and I'm, and I go, I'm so old. siara_1_01-26-2024_121542: I, I definitely, you know, remember a time before the Mac or even the, the big internet boom slightly, um, you know, slightly. Um, but I'm interested around the room. Um, do you have a significant Im, uh, memory with the Mac Mac computer or, um, what was your first experience with the Mac? I'll say mine Track 1: The see through. siara_1_01-26-2024_121542: The color, the like candy [00:29:00] color, all in one ones. Um, and everybody kind of one and you could pick your favorite color and it was just this new interesting that was like the, the hot item. You know, we didn't have influencers back then, but you definitely wanted one of these colorful Macs. jenni_1_01-26-2024_121542: that one was the movie Zoolander. Um, when the, when they, when the models were confused about how the files were in the computer. It's in the computer and I make that joke to the, like, I probably made it a few days ago because people are always like, it's in the cloud, or the handles that, or it just, it just does that, it's kind of like, oh, is it in the computer? So, um, I always think of those Maxs with that. Um, I probably have another better example, but you know, I always take it back to, uh, pop Track 1: Yeah, I think I remember being in [00:30:00] like kindergarten or first grade or something and going into the computer room and noticing the big old Macintosh boxes, what you were saying, the color you could like see through them. And then the, the Mac had the. The mouse that was just like one click or something, I don't remember. And just being like, wow, this is so cool. And it's so fresh and, and white and smooth. But I never used it because I couldn't understand like the differences. Um, but yeah, that I vaguely remember it. jenni_1_01-26-2024_121542: Bobby, you siara_1_01-26-2024_121542: about you, Bobby? jenni_1_01-26-2024_121542: back and forth earlier. squadcaster-dj4c_1_01-26-2024_121540: yeah. So, so I have an interesting story. So, believe it or not, um, they had Max, uh, my senior year at Prairie View, and I just remember, you know, being in awe because, um. I'm dating myself, but I come from a world where we had [00:31:00] typewriters and things like that and, and just having computers and just looking back now and saying that, you know, apple, you know, apple technology for a lot of people is still fairly new. I mean, it's been around forever, but we've always been Windows based and, and it kind of went away for a while. And then with the, um, they came out with the iPod, I guess it was, or the, I. Pod, yeah, the iPod is when Apple kind of made a resurgence back into, uh, into, uh, our culture and our world. So it's, I just reflect back and say that, you know, look, this is really like the first computer that I've, that I ever really worked on, and that was in the engineering lab. So, uh, those are my rec recollections of, uh, apple turning 40. siara_1_01-26-2024_121542: Awesome. Thank you. You guys for sharing. Track 1: Okay. My heat check if you've seen my background covered in chiefs. We're still in the playoffs. If you didn't know, you should know. Um, Bill's chiefs played playoff divisional game, um, last week, and of course, [00:32:00] chiefs won. Chiefs Go Chiefs. Uh, I was on Twitter afterwards. I, I, you know, every time after a game, I like to go on Twitter and just like scroll through what people are saying. And in one of the, this is not a great story, but one of the comments was a picture of Taylor Swift and she was nude. So my heat check today, and it's titled On Crunch, uh, TechCrunch, they, uh, love the Taylor Swift, uh, well puns, but swift retaliation fans strike back after explicit DeepFakes flood, uh, x AK Twitter. Um, so really, yeah. What happened was people are using AI to generate content, and in this case, they are generating images of celebrities in the nude, our, our deep fakes, which are just like taking, um. People's pictures, their faces, and then, and then kind of putting 'em on other people's faces. In this case, it was completely AI generated. Um, just everyone in the picture is not real [00:33:00] except for Taylor Swift's face and her, her likeness. And so it was just kind of an image of her kind of, yeah. In the nude. And yeah, this is starting to raise concerns. Um, and the swifties are all out and about, so they're just flooding. Um, these social, uh. Media places to make sure no one can find the images. But really this, there's a, just a major underlying concern that's, that's growing about, uh, generating ai, deep fakes and, and ai uh, non-consensual, uh, pornographic images. I, when I was researching this topic, I found out that people, this was happening in schools with, with people's children. You know, people are just taking, and it, and it goes a lot further from like a video that you create, you put on your phone, it might get hacked or something leaked. This is like people are putting, taking their pictures, putting on Instagram like anyone would do, and people are taking those pictures and then generating, uh, nude pictures from that. I, I, yeah. So we always talk about the, the pros and the benefits of ai. [00:34:00] I, I am wondering maybe have you guys thought of like this and maybe what you would want to see moving forward when it comes to like, uh, pornographic images of non-consensual people? jenni_1_01-26-2024_121542: That definitely needs to be a crime, like libel is a crime. You can't just say. you know, I don't even wanna say think out, right? You can't just like say, oh, this person is, you know, defaming, know, the black community or using bad language and it not be true. And to me that kind of falls under that same thing. You should not be, um, generating content that is a lie. I don't know how, how it could be enforced. I think that's probably the tricky part, but that, to me, that just is unacceptable. Especially if you, you know, youth, if it's in high school or even younger children are susceptible to this. squadcaster-dj4c_1_01-26-2024_121540: [00:35:00] Yeah, I, I think it, it needs to be a crime for sure, and I think it needs to have serious consequences. We've seen, um, in other areas where, where, you know, kids are getting bullied and then they go off and, and, and sometimes they, they kill themselves, you know, they, um. Uh, they can't take it. And imagine, uh, in this world that we live, if, if someone got ahold of that technology and, and, um, you know, created images of someone doing something that wasn't true, and you know, half the world's gonna believe it's that that person and half the world's gonna believe it's not. And I can only imagine the stress and the embarrassment that someone could go through. So I, it needs to be, it needs to have a very, very. Um, uh, stern punishment for doing that kind of stuff. I mean, that's one of the dangers of AI getting, getting outside of, uh, your heat check. But what information do you believe? I mean, it all seems real, so it's gonna be very hard to depict, you know, what's true and what's [00:36:00] false. Track 1: like six months ago? Oh, we're making, we're making this image and we're making an e tf, or what was it? Uh, NFT, sorry. NFT, uh, beanie Baby Fund, where we're just gonna make an image and now it's an NFT and you can pay money. I, I think that there is an actual use case that starts to come with all this AI generated content. So if it's, it's a real content, it has an NFT, you know. Associated tag to that image. And now if I'm on the interwebs and I'm looking at an image, I should be able to verify that's a real image based on that EFTI could see something like that, you know? And maybe it'll help with credibility potentially. I don't know how all that works though, so I. siara_1_01-26-2024_121542: Yeah, it's, it's insane. It's funny, I was watching a movie last night and it was, there was this like AI machine that was. warping voice voices and somebody made a statement like, the only thing that we know is real is this actual conversation that [00:37:00] we're having with people that I can touch in this room. Um, and while you know, all this technology is wonderful and it can do so many great things, there's always that flip side of it being, you know, in the wrong hands and people taking advantage of other folks. And I'm really interested. And, you know, the type of legislation that's gonna continue to come down about, um, how we're using this technology and what, punishments will be appropriate for people who are misusing it. jenni_1_01-26-2024_121542: I totally agree. Yeah, that's very unfortunate. And I wish people spent more time, spent their time on something more, um, productive to society siara_1_01-26-2024_121542: The swifties are on the case though. jenni_1_01-26-2024_121542: yeah, trying to shame people or embarrass people. That's just not good. Okay. My turn. Um. article [00:38:00] that I read it was called is from Bite, bike, go Newsletter, Netflix. What happens when you press play? So anyone that's ever interviewed with me, or if you are going to interview with me, um, I like to ask the question. I just take a random and Netflix, I think I have used Netflix before and I I just ask people, how do you think this works? Or how, if you were gonna design. Service, how would you break that down into, um, what, you know, object oriented when you're talking about object oriented programming, how would you break down the objects? How would you design this application from a services architecture? You know, what services would you need to do this, you know, to, to implement something like that? And typically, I'm looking for, um. any service like Netflix or anything you write, you have to log in. So there's some sort of security platform. Um, you gotta persist account information. There's usually like preferences or a watch list. I can't remember how Netflix calls it, but something like that where you're doing that. Um, if it's something online, maybe there's [00:39:00] comments or. Ratings and, you know, how would you organize all that information? So I really, uh, was excited to see this article about what happens when you press play at Netflix. Um, they actually, um, they, they ha are in two clouds. They use AWS for like, their, um, their typical application, like the, the app you have on your device, your phone, or you know, your, your smart TV or whatever. Um. When you're logging in and scrolling and seeing a catalog, that's all done in a AWS. And then they have a different system streaming the content and where they, you know, where they actually store the movies and the TVs, um, that, you know, TV shows that you're watching. Um, and so they go kind of go into things, um, but they. Originally, um, kind of had all their servers in-house and a long time ago, I think they said 2008 [00:40:00] was when it was kind of, I think Maddie, you mentioned this before, Netflix's 20%. What they're good at is, um, providing movies and delivering content to people, and they weren't so good at managing servers and all that stuff that goes on around there. So that's when they were like, we're just gonna outsource. That's gonna go to AWS and we're gonna concentrate. On people's experience, um, you know, watching streaming video, and this is even back when they were emailing out DVDs, I think is when they made this decision. So I always find that fascinating, especially, uh, you know, we know how we deliver software and how we are architecting it, and I always interested in other examples of how these big companies are making the, the decisions that they're making. Track 1: Yeah, you're gonna prompt us. Prompt us for a question, Jenny, so we can. jenni_1_01-26-2024_121542: there? [00:41:00] I know I didn't set you all up for a question, did I? Um, Track 1: Like, here's a bunch of information. jenni_1_01-26-2024_121542: yeah, I, now Track 1: that's great. I, no, I like how you broke that down. What, what, um, what, what was your takeaway? Let me hit you with a question. I. jenni_1_01-26-2024_121542: I, I think my takeaway was the point that, um, you need to think out. I'm in the business of, what am I in the business of managing a bunch of servers. No, I'm in the business of streaming movies. put our intellectual and our energy and our resources into making that the best experience and knowing when to say this is something that. Is just assisting us in our main business goal, but we don't need to spend a lot of resources figuring out these problems that other companies have already figured out, and that's their main purpose of the, you know, of their business. Track 1: Yeah, I agree with that. I think, you know, that's how. Businesses [00:42:00] become businesses. I I think my follow up question is always like, at what point do you know? And I feel like the answer is, you'll know, um, when you should move on. Um, but like, you know, you, you kind of always start doing things, but then when you get to that point where you're like, okay, we're, we're doing this process, or we have this, this idea. Um, but like, and we're doing it manually, but there's this company that does it really, really jenni_1_01-26-2024_121542: You gotta do that cost benefit analysis, right? You gotta be like, okay. I it, I'm gonna have to hire a team of 10 people and send them to training and make sure I have 24 7 coverage to do How much does that cost? Or, you know, versus, okay, if I outsource to this other thing, what does that cost? How is it gonna scale over time? Is, you know, you know, is this, is this a cost that's gonna keep increasing every year? Do I like the way it's increasing over, over the years? I. Uh, you, you, you, you gotta kind of put some data [00:43:00] towards that Track 1: Let's start a business. jenni_1_01-26-2024_121542: Have, I mean, in your have you done anything like that? I know I have Track 1: If I'm dealing with this, this is my cognitive load and I'm just gonna outsource this part of these chores to you and I'm not gonna worry about it. So we definitely do that in our personal life. jenni_1_01-26-2024_121542: Yeah, that's how I feel about my lawn. I don't have space for a lawnmower. I don't have, I don't wanna maintain a lawnmower, so I just ask someone else to mow my lawn I pay for it. siara_1_01-26-2024_121542: Is also. You know, you, you tend to consistently have an issue with, or it slows you down or it's really inefficient and you're like, there has to be a better way to do this and let me go find another option. Um, because this is wasting a lot of time, it's causing me a lot of frustration or I'm good at, I don't get a chance to focus on because I'm spending so much time on this. This other thing that I'm probably not very good at, is definitely a use case. For that. jenni_1_01-26-2024_121542: Yeah. [00:44:00] Yeah, and I, I think one of the fundamental principles of software development is like, don't reinvent the wheel. So especially if you're starting from scratch. You're kind of like, oh, you know, I need a, you know, I need a library to, uh, you know, manage, um, database connections. Well, there's gotta be a zillion out there. You're not gonna build that from scratch, so you know you're gonna use something that exists out there. Yeah, yeah, yeah. Don't bog down your life with that if you know if there's another way to do it. siara_1_01-26-2024_121542: Thank you Bobby, for hanging out with us today. Where can people find you out on the interwebs? jenni_1_01-26-2024_121542: Awesome. siara_1_01-26-2024_121542: Awesome. ​[00:45:00]