(0:00) what they need, how they need to solve their problems when they use software, how they work, (0:08) what workflows do they expect, and all that good stuff. And that's a very UX-focused job, (0:14) and that's called requirements. We're going to talk about that in week two. And then (0:19) then got into the operational side of delivery of product teams and fell in love with product (0:33) ownership and product management and kind of never looked back from research to design to product. (0:40) Yeah, Marie France, thank you for calling that out. I think we still have time. So (0:48) I'd like you all to get to know me a little bit. And I've never really looked back from (0:55) mixing all together. And 10 years later, I've kind of gone back and forth between (1:04) research and design and product management, scaling teams and products and working on (1:13) probably teams that have multiple products at once. (1:18) How do you do that? How do you orchestrate that? How do you build operations for that? (1:25) Worked on companies that are large and small, but even small companies can be enterprise companies. (1:31) It just depends on what they're building and what they're doing. And so I've worked on a lot of (1:36) enterprise products before. Of all shapes and sizes and all different types of (1:45) purposes and uses. And I think that this is such an important class, (1:56) because as a person who is orchestrating the product, (2:02) you have to be able to orchestrate all of these areas without necessarily knowing (2:11) what the production entails, but you're building owners in the work. (2:16) So you're building owners in the development, you're building owners in the technical architecture, (2:20) you're building owners in the design work, you're building owners in the research work, (2:23) you're building owners in the business work and the sales. (2:27) And that's really what makes a good product manager. That's what makes a good (2:33) leader of product teams. And this is a great age of being able to do that. (2:43) Being able to learn just through talking to AI models is such a powerful thing. (2:51) We're going to do that a lot in this class. This class is a lot of self-discovery, (2:57) I would say, through using the tools that you have with you. (3:03) Never really forgetting your critical thinking and your brain in the meantime. (3:13) We're going to have five hours a week together. We're going to have actually seven hours a week (3:17) together. We're going to have a three-hour session on Saturday and repeat lecture is going (3:24) to be on the morning tomorrow. So you don't need to come to the repeat, but you should come to the (3:30) working sessions because they're going to be different every time when we work together. (3:39) We're going to have the basically progress in the work. So we're going to decide the session (3:48) of the work in the working session. And it's going to be a continuation based on whoever's (3:54) in the room. So if people weren't in the room before in the previous session, I'm going to (3:59) catch them up. And then we're going to keep going on the work based on the topics. And that's how (4:04) we do working sessions and classes that I run. So if you miss a working session, I would highly (4:10) recommend watch the recording. You will catch so much in that recording. If you don't have an (4:18) ability to make the working session, it's okay. The ones in the week are one hour and the ones (4:24) on the weekend are three hours. The three-hour session on Saturday, we're going to dig really (4:30) deeply into whatever the topic is. This week will be later than normal weeks. And starting next (4:41) week, we're going real deep with deep rigorous topics every week. Yeah, drop your answers in (4:50) the chat. Okay, everybody click on that link in the chat and please follow me in FigJam and I will (4:59) take you on the tour of the basics. You probably read the syllabus. I'm not going to bore you with (5:10) reading the syllabus again. But I do want you to know that this is, I'm going to treat this like a (5:19) master's level education where I am going to probably scaffold more than I give you answers. (5:29) And, you know, I have a lot of knowledge. I can share a lot of knowledge. But I don't want to (5:34) build a dynamic where I'm the master teacher and I have all the answers and I'm all-knowing and (5:40) I'm the authority in this. I want us to be able to have a discovery together with all of these (5:48) topics that we're learning. Because I'm not going to tout myself as the 40-year expert that knows (5:53) everything in my head and I'm not going to build a hierarchy between teacher and student like that. (6:02) But I am going to hold you to a lot of deep expectations about putting in a lot of commitment, (6:06) putting in a lot of time, putting in a lot of self-starting. And there's going to be a lot of (6:15) I don't know what I'm doing, so I'm going to try to figure it out. There's going to be a lot of (6:19) troubleshooting in this class for you. There's going to be a lot of you dealing with issues (6:27) or errors or bugs. And one of the things that makes you a great (6:35) leader of the work is the ability to critically think through troubleshooting (6:40) and to resolve unknown issues that happen. And so nobody can really teach you that. (6:48) We can walk you through it. We can't prepare you through that with a lecture. I want to focus on (6:56) discovery and doing troubleshooting. You all being able to get into get your hands dirty as much as (7:03) possible. So all these topics cover the different parts of what makes enterprise-ready products. (7:13) And I'll tell you what the definition of that is. (7:18) So this week, we'll go into introduction stuff. Next week, we're going to talk about (7:24) I'm going to give you an entire course of product management in one week. (7:28) And you're going to be producing an entire spectrum of deliverables for your coding challenge (7:37) next week for product management so that you can get into week three, (7:42) starting to build what's called a three-tier application, data-driven hosted web app. (7:54) Week four, we're going to get into the deep rigors of engineering data, (8:00) setting up the data structure, setting up the organization, setting up the pipelines, (8:05) using that data. Sounds real fancy. You're probably going to learn it and you're going to be like, (8:11) that's it? That's all it is? It goes deep, but it's not that complex. I think what makes it (8:19) complex is the problem solving and maybe the definitions of how all that stuff works. (8:26) Week five, we're digging really deeply into security. Again, an entire course of (8:35) information technology and information security and cyber security is going to get squeezed into (8:41) one week of work. And you're going to learn a wide swath of things that relate to how do you make (8:50) your application very secure? That's an extremely hard answer to answer. (8:58) I have a whole degree around that. And I'm going to shove it into one week of (9:03) learning and doing and practice, and you're going to learn it. (9:10) Week six, I'm going to take an entire course on modern software architecture, and I'm going to (9:17) shove it into one week of learning. And you're going to produce things that are actually scalable (9:26) in nature. When I say scalable, I mean that millions and millions of people can use that (9:32) without the software breaking down, becoming way too slow, and actually giving people a good (9:39) experience at scale. In week seven, we're going to talk about how do you connect different (9:45) applications and agents together? Sounds fancy. It sounds really foreign to you, probably. (9:53) Maybe, maybe not. But it's not that fancy. Once you understand how the web works, (10:00) I will start with the basics tonight. That's the foundation that you need to understand how (10:07) all this works. They got really, really fancy words, but you don't really need to (10:15) think really hard to get the concepts. If the concepts are broken down to you, (10:20) anybody can get them. And then in week eight, we're going to talk about how do you make (10:27) that thing that you create visible, scannable, public on the web? (10:33) Either people are going to ask LLM models about something, and it will recommend your product. (10:42) It's called Ask Engine Optimization, or they'll use Google Search, or Bing, or Yahoo, or (10:50) Brave, or whatever. And those will pull up in search results. That's called Search Engine (10:59) Optimization. And that will come right to the end of probably the beginning, the beginning slash (11:10) middle of December. We're going to take a little bit of a break for the holiday. (11:16) We can talk about what we should do in between that after the first eight weeks. The second (11:22) half is going to be us, all of us joining project training teams in TechFleet. And if you want to (11:31) bring an open source application into the mix, we can start an open source application team. (11:35) There's about seven or eight client projects that will be going on at the same time. (11:42) Maybe not like right at January 1st, but right at the start of the new year that (11:50) you will be able to join. We're probably going to have maybe teams of four to five developers, (11:57) and other people will be on the team to work on product management, research, design, (12:04) project management, product strategy. And they'll work with non-profit clients. Either TechFleet (12:10) will be the non-profit client, or we're going to have other non-profit clients, and you'll work (12:15) with them, that team, for a project phase, which is 12 weeks typically. It's eight weeks of project (12:24) work, and we do onboarding for three weeks. So it's eight weeks of actual sprint work that you'll do. (12:34) Raise your hand in Google Meet if you are new to TechFleet. You've never done a project in (12:38) TechFleet. Raise your hand. Morgan, repeat the question. I stepped out for a second. (12:58) Yeah, no, no problem. Who has never been on a TechFleet project before? It's not you, Jalon. (13:03) Right. Cool. It's going to be a pretty exciting time for you all. If you've never been on TechFleet (13:12) projects before, you'll get trained. And raise your hand if you have been on a TechFleet project before, (13:21) so I can see if you're awake. Okay, now the question is, who didn't raise their hand? (13:35) Lots of people have taken a TechFleet project before. That's really awesome. (13:41) I think you're going to be coming out at a completely different level after this (13:45) residency. I'm not even going to call it a class. It's not really a theory classroom, (13:53) not in the least bit. Raise your hand if you joined TechFleet this year. Okay, (14:13) raise your hand if you were in TechFleet for more than two years. Wow. Raise your hand if (14:28) you've been in TechFleet more than three years. Wow, you all are the OGs. So Joe, (14:42) were you on projects, Joe? I wasn't on projects, it's funny enough. I was on the Slack. (14:52) Yeah, so I logged into the Slack and I was like, hmm. Wow. Yeah, so it's however long that's been. (15:02) Yeah, I would consider you one of the OGs of TechFleet. That's great. You're welcome. (15:14) And also, I want to get a sense of what kind of roles people have studied before. So, (15:22) raise your hand if you have studied and want to be in UX. Any area of UX, raise your hand. (15:39) Holy moly, most of the class. Okay, raise your hand if you are studying to be or want to be (15:47) product manager, something similar to that. Cool. Okay, how about developer? Raise your hand if you (16:04) have or trying to study to be developer. Cool. Amanda, is it official now? Developer? Is that (16:21) you're going for? I've picked all three so far. All right. Who has no idea? Who was like, I'm (16:33) still figuring it out. That's okay too. This is a great place to do it. TechFleet, other communities (16:41) like this. Cool. All right. Well, we've got mostly, how about project management? Anybody (16:53) in project management or ops trying to get in, get into that field? We've got a wide mix of people. (17:09) It's great. I think that the beauty of this era of like, what does the future hold is, (17:19) I think the lines are blurring. And I think that things are going to become, (17:29) I think, a little bit more defined by us over time. I think that there's going to be a lot (17:38) more business owners over time. And you can literally take the skills that you're (17:44) practicing in this residency and go build a business after. And I think that that's where (17:54) that's where it gets exciting for some of us. And probably scary for a lot of us too, (17:58) because some of the skills that we haven't, that we were training for before haven't necessarily (18:04) been the ones that would get you hired these days. But we're going to put our heads down and (18:13) learn enterprise software development, engineering, enterprise engineering. That's what we're going (18:19) to learn in this class. So I've got all the links here in one place. Bookmark this class. (18:25) Bookmark this Figma page. All your content is going to be on this page here. If you zoom out, (18:34) there's week one lecture. I'm going to go into it in a second. And the link to the syllabus is here, (18:42) the link to the working sessions, or sorry, the courses that you should finish before you get (18:51) onto Teams. They're in the TechFleet platform. You should finish those before January. (18:58) You should do these things. You should, you know, there's all the basics, all the logistics. (19:07) The weekly schedule is here. The coding challenges are here. Click on these links. (19:13) There is a helper bot, an AI bot that can help you finish some of the deliverables, (19:20) but not all of the deliverables. And mostly it's stuff from week two. (19:28) But if you want the link, here's the link. And premium residents, raise your hand (19:36) if you're on the call and you're a premium resident. Well, fill out that form. This is (19:45) your advanced project intake form. Y'all are going to pick an advanced topic. You're going (19:51) to work with me 10 coaching hours, and you're going to do an independent study on that topic. (19:58) And then you're going to do an advanced project. That's not the coding challenge. It's like same (20:03) topics of like learning, but different application. So fill that out. It's okay if you don't know, (20:12) just fill it out. And there's a partner that asks, do you know, or do you not know? (20:17) And I've accounted for that. Okay. Any questions? We covered the basics. (20:27) I'm going to go into deep lecture here. One quick question. Yeah. When's the last (20:35) like day to be able to upgrade to the premium account? (20:42) Linda, you broke up when you asked me that. It might be my internet. Can you repeat what you said? (20:47) Oh, sorry. Can you hear me? Yeah. Yeah. You're good now. (20:51) Okay. I was asking when would be the last day to be able to upgrade to the premium account? (20:57) Oh, okay. There's it's not necessarily a day. There's one spot left. I have to limit my time. (21:04) So I've got one spot left as of, I think there's still, yeah, there's still one spot left. (21:14) Just email me. Anybody who wants to like already, you already got into the class. If you want to like (21:20) have those 10 coaching hours, then email me. I'll send you a link. Okay, cool. (21:33) Any other questions? No. Good. Yeah. Machia. Sorry, really quick. You said that we're going to (21:50) be working on workshop from Saturday, correct? Yeah, it's going to be a three-hour working (21:55) session. It's not a workshop. It's more like we all get together and we work. We're going to (22:00) probably do breakouts some weeks. We're going to code. It's going to be coding like AI development (22:08) or whatever the topic is that week. Okay. So which will be part of our homework as well, (22:14) correct? Yeah. Something different. Okay. Yeah. So I'm going to code on examples in class (22:23) and then the homework is on you outside. So you get that double learning. You get to watch me do it (22:31) or watch other people do it in the working sessions and then you also apply it to your (22:35) homework. Okay. Got it. Thank you. And that'll be recorded as well? Yeah. All of them will be (22:40) recorded. I'll put a page together on Notion so that everybody has one place to go to get all the (22:48) recordings. Okay. Great. Thank you. Yep. Anybody else? All right. Here we go. Follow me. We're (23:08) starting. We're going to start real simple. I like this GIF and this GIF is great too. (23:17) What's the difference between vibe coding and actual engineering? Well, if you look at these (23:23) memes, you can kind of get a sense of what vibe coding is, but I've broken down a definition for (23:28) you. Vibe coding, in my own words, is if you type something into an AI prompt and you hope for the (23:34) best and you let it do all the decisions and you let it do all of the work behind the scenes and (23:41) you don't check its work, you don't actually orchestrate any of its work, (23:46) here's where it's risky. It's risky because it creates very hackable applications. It could (23:54) very easily expose user data. It could make bugs, can make a software completely unusable. (24:01) It creates a really bad experience for people. It's not a good idea to do this for a business (24:07) and it'll expose private data. When you're vibe coding, you're typically just describing what you (24:14) want and using tools like Lovable or whatever AI tools have been hyped to create an application in (24:25) two minutes. That's what it's doing. It's doing a bunch of decisions behind the scenes that you (24:30) think about, but there are obviously consequences to that. It's not good for making products that (24:39) actually people would use. It's not good for scaling a business. It's not good for making (24:44) a product that people would pay for, but it is good for rapid ideation, coming up with ideas (24:50) really quickly, making concepts that maybe you want to throw away. Great for that. If you get (24:57) into engineering, what we're going to call enterprise AI development, you are following (25:08) all of the standards and rigors that make applications able to be used by millions of (25:13) people and perform well and not get hacked. It takes a lot of rigorous workflows, way more (25:24) workflows, way more processes, way more moving parts, not necessarily a risk. It's more like a (25:32) challenge. It takes more time to produce and it will cost you more credits when you're doing this (25:39) with AI development. You're using a lot of tools that aren't AI based at all. You have to create (25:52) the definitions of software patterns. I will teach you that. You will have to create definitions (25:57) of technical architecture. You have to secure your coding features. You have to engineer data. (26:04) You have to build your hardware and design that structure in such a way that it works well. I'll (26:12) teach you all this stuff and you have to test for security and you have to test for quality. (26:19) I'll teach you that too. A lot of the tools that you use aren't AI based at all whatsoever. (26:27) They're the tools that teams have used for 40 years. You either have a human developer or you (26:34) have an AI LLM as a developer and the rest of it is human knowledge, testing tools, database tools, (26:44) web hosting tools. It's not really great for building a quick concept that gets thrown away (26:52) because it's probably too much, probably too expensive, but it's really great for building (26:59) real code and building bug free software. It's great for building applications that can be used (27:05) by billions of people. It's great for supporting teams of multiple developers and maybe cross (27:12) functional people too. It allows you to comply with all these regulations that are out there (27:20) when you build products that people use or buy. There are a lot of benefits to it. (27:27) And in the spectrum of building things, all these circles in green are the areas that you (27:36) are going to learn in this class. You're not going to master them. You're not going to memorize them (27:42) but you don't need to. You just need to practice them and you have to have an awareness of where (27:48) they all fit into the bigger picture. So the things that we're going to cover, data engineering, (27:55) data analysis, product ops, quality engineering, requirements, info security, dev ops, network (28:02) technology, development, technical architecture, security, and I would say also designing (28:12) agents as well. And I mentioned this before, I think that the future of what we do, (28:19) when I say we, I mean people that build products and build services and make change in the world, (28:27) there's going to be a lot more multidisciplinary stuff around it. There's going to be a lot, (28:32) there already is, there's going to be a lot of shifting the ways that things have worked before (28:40) or making a decision to go back to the way that things were before because the tool that we are (28:45) now introduced to makes it worse for everybody involved or the consequences of it are too extreme (28:54) and we're going to end up saving a lot of time when we do tasks. We're going to end up (29:00) doing a lot more work than we were trained in, which is why you're here. We're going to work on (29:07) a lot more strategy and operations and requirements and orchestration of work. We're (29:14) going to be doing a lot more relationship building and influence building. We're going to be, (29:20) we're going to all become end-to-end experienced designers in our own right, (29:27) systems, designers of systems. There's going to be way more business problem solving. (29:33) You're going to need more knowledge of the full stack development, which is why you're here. (29:37) You're going to learn all that stuff. There's going to be a lot more automation, but that gets (29:42) a little complicated and it's not going to be like automated the whole way. And there's going to be (29:48) different types of UX. There's going to be different types of user experience to build in the future (29:57) that we've never really seen before. We're at the cutting edge of it right now. (30:02) Building agent models and agentic applications. And I know that this is an AI class, but I want (30:12) you all to promise me this. Come off mute and promise me that I will not replace my brain (30:21) with AI. Come off mute and tell me that. Everybody say that. Repeat after me. (30:28) I promise not to use my, replace my brain with AI. (30:35) Say it again. I promise not to replace my brain with AI. (30:50) I heard somebody laughing. Say it one more time. I'm serious here. (30:53) I promise not to replace my brain with AI. It's so important. Don't do it. Repeat after me. Go. (31:01) Say it one more time. Okay. Great. Okay. We're going to build a bunch of critical thinkers in (31:13) this residency. Don't replace your brain with AI. Don't do it. Don't offload your thinking. (31:22) Don't trust AI's judgment. Don't trust AI at all. Build workflows where you can verify (31:29) everything. Don't stop giving it your own data. Don't let it make the decisions for you. Don't (31:36) stop learning the fundamentals. That's why you're here. So you can orchestrate and verify AI more. (31:42) But you should ask it to compare your results to authoritative stuff. You should define the (31:49) outcomes that you want to achieve with it. You should provide context. You should ask it to (31:53) provide observable proof. You should ask it to teach you the steps as you're doing it. And you (32:00) should ask it to skeptically audit everything that it does. When you do that, you will be able (32:11) to rigorously extend your capabilities of everything that you do. When you don't offload (32:20) your thinking to AI, this is what you need regardless of anything with AI. Forget the tool. (32:28) You need a solid foundation about engineering, security, and operations. You need your brain, (32:39) critical thinking, and problem solving. You can't ever lose that. You need a system (32:49) to create and track your documentation and your specifications. That isn't going away at all. It's (32:57) more important than ever. You need great organization. You need to think like an agile (33:02) person. And you need great people skills. That's what you need to be a really awesome AI engineer. (33:10) What you're going to learn in this class has everything to do with the whole end-to-end cycle (33:17) of product development life cycle. If you're in tech fleet, you know this. You know this flow (33:24) chart. This is the bedrock of what we do in tech fleet. We build and ship products and services in (33:32) the world. And the things in green are the deliverables that you're going to learn in this (33:39) class. The thing called a milestone is a chunk of progress. And each chunk of progress has (33:47) a workflow, sequences of things, and then you produce each thing, and then you keep moving, (33:52) and you keep producing, and it happens in a cycle. It happens in some type of never-ending routine (34:00) where you are discovering and exploring and making change and then validating that change, (34:07) and you keep building and building and building and building, and you keep understanding what (34:11) you're working toward. This is how we think when we operate in an agile way. You will never lose, (34:28) no matter what you're doing. Agile is not a series of methods. Agile is a philosophy. It is a way that (34:35) you think and a way that you work. And to be really great at what you want to do, (34:44) you're going to be really great at thinking like an agilist. Here it is. We are uncovering (34:52) better ways of developing software by doing it and helping others do it. And through this work, (34:58) we have come to value individuals and interactions over processes. We talk it out with people. We (35:05) don't need to know what the process is. We just talk it out with people. We deliver working (35:10) results over documenting comprehensive results. So we ship it into the hands of people, and we (35:17) learn from it, and we don't worry about documenting how it works. We just ship the working result. (35:24) We collaborate with customers over negotiation of some type of contract. We don't make a giant (35:31) doc in the beginning that says, this is everything you're going to do and nothing more. (35:35) We negotiate with people as we go. We work with humans to agree to that as we go. We might change (35:42) our minds. And lastly, if we change our minds, we're going to respond to change. We're not going (35:53) to be married to the plan that didn't work just because we had a plan for it. And we're flexible. (36:01) That is, while there's value on the items on the right, we're going to value the items on the left (36:06) more. And everything that we do, everything that we approach as teams who are thinking and working (36:15) in an agile way, we just completely change the ways that the work happens. We might start with (36:22) zero understanding and ship and learn based on what that understanding needs to be. And over time, (36:30) as you keep cycling through that, you learn more and more, you deliver more and more value. (36:37) So this was the image that everybody used to share. If you don't build the thing in parts, (36:44) you build the most important thing, and then the next most important thing, (36:48) and the next most important thing. And then this piece of garbage came out. And (36:54) I don't believe in this notion at all, whatsoever. You're not building the most complex thing first, (37:03) and then reducing yourself to the least complex thing. If you want to build proper enterprise (37:09) engineering, you're not abandoning agile. You're not abandoning the way that you think about (37:15) responding to change, delivering working results, collaborating with customers, and talking to (37:22) people, instead of building perfect processes. You're building the most important thing. You're (37:28) learning from that most important thing. You have an idea, you validate the idea. You have another (37:33) idea based on that idea, you keep going, and you validate, and you validate, and you validate. (37:37) And that might look like something that, whatever that version is, might just be faster to do. (37:47) That's still going to be the most usable, valuable, testable, minimal piece of delivery (37:58) at any given time. So, we're not abandoning anything, whether it comes to agile. And agile (38:06) is not a method to abandon. It's not a series of steps. It's not tools. It's the way that you think. (38:14) It's the way that you work. If you are working and thinking in the ways of waterfall, you're (38:20) thinking in departments. Everything gets defined up front, then you do all the research, and then you (38:26) do all design, and then you do all the development, and you don't talk to anybody. You just finish what (38:31) you agreed to, and then you release it all the way at the end. That's not the way that we agilists think. (38:39) We think in, let's get it out. Let's collaborate with each other. I want to hear from the non-developers (38:45) in the room what the risks are. I want to identify together the most important and the riskiest (38:52) things that we need to de-risk, so that we can work quickly to get it out into the world, working (38:59) results, and validate that. And then figure out, does our plan need to change? And that's how you do it. (39:08) That's how you build things that scale and grow and maintain themselves in the world. You end (39:14) up doing this work in parallel. That's not because we introduced AI into the workflow. (39:21) Guess what? We were doing that 30 years before, too. It's called agile. We could ship something (39:27) in a week, too. Great. It's called following an agile philosophy. So, be careful what you hear (39:35) and what you take away from what you hear, because none of this stuff is going away. (39:44) The only thing that's changing is you have a human tool or you have a robot that learns as it (39:55) produces things. And both of them are going to be prone to mistakes. Both of them need context. (40:03) You need to be able to still orchestrate, discover, build vision, build scope. You need to (40:10) intake the work. You need to work with others in the work. You need to collaborate with people. You (40:14) need to build influence. None of a, like any one single developer, whether it be human or machine, (40:21) is going to do that for us. Ever. And the challenge of this is not the speed of it. (40:32) It's the alignment of it. We're going to talk about that week, too. You easily will be misaligned in (40:39) everything that goes on and could be orchestrated. So, AI development isn't just about (40:46) development. It's about the process of making change in the world. It's about following this (40:52) life cycle. You're introducing AI as a tool to produce those things at different steps. (40:59) And that's it. And once you learn the tool and how to do the tool, you're good. You can build (41:09) entire products that actually work. But oftentimes you won't be able to do it alone. (41:16) You'll still need people. You'll still need cross-functional teams and people working (41:22) together. There's part one. Who's got questions before I go into part two? Part two is going to (41:31) be quick. What questions do you have? Stop me if you have questions. Okay. Now we're going to get (41:43) real technical. Here's how the web works. If I go to Amazon.com, and you need to know this (41:52) so that you know the fundamentals of how to build software on the web. You need to know this. (42:01) This has nothing to do with AI. This has nothing to do with the technology tool. It has everything (42:06) to do with how the web talks and the language that it uses for one computer to talk to the next. (42:13) If I go to this website, there are countless different things that are happening to load (42:23) this. Where does this all come from? All these things right here. Where does this all come from? (42:30) You have to understand where it all comes from so that you can build that (42:36) and develop and engineer that. If I type the URL in my browser on my computer, (42:46) it will develop what's called a request in the language of the web, which is called HTTP. (42:57) Hypertext Transfer Protocol. Fancy, fancy words. HTTP request looks like this. (43:06) It has certain standards that you have to follow. It has information. What you're doing is you're (43:15) requesting all the information that is on someone's computer, Amazon.com. I'm going to make (43:24) the request on my browser. It does it for me. It does it in a split second. Then it's going to (43:31) travel the interweb, the internet, all the little tiny servers and data warehouses and (43:38) all the things in the world. It's eventually going to hit Amazon's computer called a web server. (43:47) A web server hosts files and images and scripts for a website. Any website in the world has a web (43:59) server. A server is a type of computer. Those things take requests. There's a request that (44:07) comes in. It's like someone's knocking on your door. Knocking on your door saying, (44:11) hey, I want to visit Amazon.com. Please give me the files. (44:16) Then Amazon.com web server has scripts that it stores all the files. These things need to be (44:24) secured. There's coding involved on this. You make that as a developer. It's going to then (44:32) work with another computer that is on Amazon's side called the database server. The database (44:40) server only has a bunch of data. It's stored. It is like a chest. You open the chest. You see (44:52) what's inside the chest. You can pull out gold. Then you take it away and close the chest and (44:58) that's it. These two computers talk to each other. One has a script to say, hey, database server, (45:06) give me this data. We're getting requested to have this data, so I need this. They do what's (45:14) called a query. The database either says, okay, great, you can have that or no, you can't have (45:22) that. Each thing has two parts. It has a request and a response. The web server has a request (45:31) to the database. The database has a request, receives that request, and then says back to (45:38) the web server, here's your response, failure success. Here's the data. (45:46) And then the web server says, thank you very much. I'm going to now send a response back to (45:51) my browser. Now the user who has requested this website is now going to receive all the files (46:00) in my browser, goes back on the internet, back on this crazy thing called the cloud, (46:07) through a bunch of servers, through a bunch of computers, back onto my computer, (46:13) and in the split of a second of an eye, this thing returns, this is what it looks like, (46:20) this is an HTTP request. It returns web files, images, all the other stuff, scripts on the (46:29) browser. When you develop web applications, you're developing all of this. You're either (46:36) developing an application with a database, or you're developing an application without a database. (46:44) And give me an example of an application that doesn't have a database, a two-tier (46:53) architecture. Somebody just shout out an example of two-tier architecture app. (47:09) Think critically. Don't lose your brain to AI. Would APIs count? APIs are like an HTTP request. (47:22) They're just like language. Probably most marketing websites, they're pretty much just (47:30) a brochure. Yep, yep. So if I go to TechFleet's website, TechFleet's website doesn't have, (47:38) this website doesn't have a database. It just has, it has files, it has scripts, it has (47:46) a web server, but it does not have a database, because it's not storing data. Yeah, good job. (47:53) Give me an example of a three-tier architecture that has an application with a database, (48:03) like Spotify. Sure, yeah. What's the data in Spotify? What data is it storing? (48:10) Is there music, mp3s, WAV files? Yep. Give me another one. One more. (48:25) Like any bank website, right? Sure. What's the data in a bank website? (48:31) What data am I pulling in a database? All the information that is pertained (48:44) to the user, like their monthly bills and stuff like this. Yep, good job. Y'all got it. (48:57) We're gonna build this right here. We're gonna build, in this class, three-tier architecture (49:04) architecture applications. That's what you're gonna build. You're gonna build the server. (49:12) You're gonna build the data server and the web server. You're going to secure the web (49:17) server and the data server and the client. You're gonna learn how to do that, and you're gonna learn (49:22) how the requests happen between the web server and the database server, and an agent is running (49:29) on a web server, and an agent needs to talk to a database server. The agent might live here. (49:37) You might have multiple web servers. You might have different things that do. You might have a (49:42) server that sends an email to people, and you might have another server that runs an agent, (49:47) and you might have another server that hosts the website. You're gonna learn how to build all that (49:53) stuff from all these tools that we're gonna do. I also need to show you how (50:00) Clod works. We're gonna use Clod. We're just gonna have class for like two more minutes, (50:05) and then you're good. When we go into the working sessions this week, you're gonna learn (50:11) how to set up Clod, and it's not that. It's very simple. You're gonna learn how to build skills, (50:18) and you're gonna learn how to set up GitHub, so you're gonna download Clod on your computer, (50:25) which only runs on the computer. It does not run on a web server. It does not run on anything. It (50:31) only runs on the client, so Clod in this class is only the client, not a web server, (50:42) not a database server. Your computer is where it's all stored. That's it. You're gonna build (50:50) applications that have servers, but Clod desktop is only on your computer. You're gonna build a (50:56) new session. Every time you do a new session, it starts an AI bot, so if you have four sessions (51:04) going on at once, here's an example. I have seven different coding features that I'm working on (51:12) right now. They're all in different states. All of these are agents that are spawned anytime you (51:20) hit new, and those agents have memory. They have files that are like regulations. They're like (51:32) its brain. They'll look at the files anytime it starts, and it will rely on those files (51:41) to know what to do and how to build things and what not to do, and you can build these (51:46) files to not have to always instruct the AI to do the same thing. If you find yourself using the AI (51:54) and telling it over and over and over again, stop using emdashes, you could build a file that serves (52:01) as its memory to stop that from happening because it looks at that and it's like, oh, I shouldn't use (52:06) emdashes. That's called a skills file. This is a relatively new field, a new area with AI, (52:17) but it's kind of evolving, but it's not that crazy to do. You just have to know (52:23) what the structure of it is and where to put it on your computer, so it's relying on that. It's (52:28) memory. It's like the brain of the AI, and then there's also files on your computer, (52:36) and it's looking at the files on your computer because it's only living on your computer, (52:41) so when you download code, you're going to put it on your computer. (52:47) It's going to make the change on your computer, and then it's going to push it back to the cloud (52:54) called GitHub, which is where your code lives if it's been published, and they can also talk to (53:02) the internet, but this is kind of how the AI is going to work when you're working on cloud desktop, (53:13) so those are the basics. I'm a little bit over, but this is like the very beginning of everything. (53:20) I did record this, so I will share it out if you all need more reference or a refresher. (53:31) You can always watch this. I'm good to stay for questions, whoever has questions, (53:38) and when we get into the working session this week, I'm going to go into the stuff that's (53:42) in the green boxes, read it on the FigJam file, but for now, what questions do you all have? (53:51) What are you curious about? I just have a good question. (53:55) Morgan, if you could please do me the favor, send me that link to upgrade, just a reminder. (53:59) Yes, yeah, I will. I think I sent it to you before, Lando. It's actually the same one as before. (54:06) Oh, okay, perfect. In your email. Yeah, same link. Let me know if you can't find it. (54:15) Are we by chance going to work with any agentic OS? (54:21) Well, so if you do premium, you can pick that as a topic. I'm going to show you (54:28) how to build agents from scratch. Okay. Like when I say from scratch, I mean using an API, (54:41) like some platform that you're, it's not using a tool that's a third-party tool, (54:47) you're actually coding it on your code base, how to start an application that uses an agent. (54:56) Okay, when will we get the cloud accounts? (54:59) I sent the emails to those in your, you should have those in your inbox (55:05) for everybody. Everybody check your inbox. You should have an invitation from Anthropic for (55:12) your seat to cloud, and it should be under Techfleet's organization. (55:25) I don't think I ever got one. (55:29) No, check your spam folder and Lando, if you don't see it, email me. (55:37) I'm going to be able to resend you the invite. (55:42) I'm having a really hard time getting into Discord. I had an account (55:46) long time ago associated with a different phone number. (55:49) Oh, no. (55:51) Yeah, now it's taking me all kinds of trouble trying to get in there. (55:56) Yeah. (55:57) Is there anything critical that I need, like kind of immediately or is it optional? (56:04) It's for chatting with like everybody in the class. (56:11) I will give you a link to the support form (56:19) for Discord. Usually you have to submit a request for them to delete your phone number (56:27) and help you through it. I used to have to help people do that when we first moved into Discord. (56:40) Yeah, so like if you have that existing phone number on an old account, (56:47) they have to remove it first before you can finish your other account. Or if you (56:53) are able to log into your old account, they'll usually try that first with you. (57:00) Okay, cool. Thank you. (57:02) Yeah, but you're good until then. I'll just keep joining these meetings and anybody. (57:07) There's like one other person who's not able to get into Discord. (57:09) For the first two weeks, I'll probably email people so that nobody misses anything. (57:21) What else? Throw anything at me. I'm here for any questions. (57:27) Would you like us to do this week one homework before the working session? (57:31) No, you're good. You can submit it next week. (57:34) Cool. (57:35) You can finish it now if you want. Like you can finish it as soon as you get in the cloud. (57:48) Cool. How the web works, did that help you all like understand how these applications are built? (57:58) Well, that's good. Yeah, once you understand, that's what they train you in IT. They train (58:11) you on these like seven layers of the network. And once you understand that you understand exactly how (58:16) to fix problems and code things, it'll make you a much better producer of the work. (58:24) Whatever you're doing, if you understand this stuff. (58:29) Cool. All right. Don't come to the repeat lecture. Unless you want to hear it again, (58:35) you're welcome to if you want to, but don't spare yourself otherwise. (58:40) Just the same thing. And thanks for a great first lecture. (58:45) I'm looking forward to this class. I'll see you all in the working sessions. (58:49) Thanks, Morgan. (58:50) Yeah. (58:50) Thank you, Morgan. (58:52) Thank you. (58:54) Have a good one. (58:56) Have a good night.