WEBVTT 00:00:00.000 --> 00:00:02.640 You know, I was looking back at my old college 00:00:02.640 --> 00:00:05.339 curriculum the other day. Oh, yeah. Not for nostalgia, 00:00:05.480 --> 00:00:08.500 really, but just curiosity. And I saw Introduction 00:00:08.500 --> 00:00:11.320 to C++ icon there. Ah, the classic. I remember 00:00:11.320 --> 00:00:13.400 spending two entire semesters just trying to 00:00:13.400 --> 00:00:16.480 understand the syntax. Years of my life, honestly, 00:00:16.699 --> 00:00:18.519 dedicated to just getting a computer to print, 00:00:18.940 --> 00:00:21.059 hello, world, without some fatal error code. 00:00:21.120 --> 00:00:23.019 It was a rite of passage, wasn't it? The syntax 00:00:23.019 --> 00:00:25.120 was the gatekeeper. You couldn't build anything 00:00:25.120 --> 00:00:26.719 if you couldn't speak the machine's language 00:00:26.719 --> 00:00:30.530 perfectly. Exactly. Looking at the deep dive 00:00:30.530 --> 00:00:33.689 we've prepared for today. I genuinely had this 00:00:33.689 --> 00:00:37.369 moment of I don't know existential shock We're 00:00:37.369 --> 00:00:40.969 looking at a workflow right now in 2026 where 00:00:40.969 --> 00:00:44.350 that entire barrier the years of syntax the missing 00:00:44.350 --> 00:00:46.969 semicolon nightmares It's just it's gone. It's 00:00:46.969 --> 00:00:49.189 gone. But what's so fascinating is what replaced 00:00:49.189 --> 00:00:52.369 it the bottleneck didn't just disappear It moved 00:00:52.369 --> 00:00:55.409 it shifted from can you write the code to what? 00:00:55.509 --> 00:00:57.570 Can you think clearly enough to ask for it? That 00:00:57.570 --> 00:01:00.280 is a massive pivot So the only thing stopping 00:01:00.280 --> 00:01:02.719 me from building software isn't a technical skill 00:01:02.719 --> 00:01:05.459 anymore. It's just my own ability to articulate 00:01:05.459 --> 00:01:07.900 what I want. Precisely. Clarity is the new code. 00:01:08.099 --> 00:01:12.420 Okay. Let's unpack this. Welcome to the deep 00:01:12.420 --> 00:01:16.200 dive. It is Tuesday, January 27th, 2026. Yeah. 00:01:16.359 --> 00:01:19.739 I'm here with MacGuy co -host and today we are 00:01:19.739 --> 00:01:22.379 getting our hands dirty. We are. We are just 00:01:22.379 --> 00:01:25.120 talking about AI trends in the abstract or what 00:01:25.120 --> 00:01:27.879 might happen. in five years, we were looking 00:01:27.879 --> 00:01:31.000 at a concrete step -by -step workflow for how 00:01:31.000 --> 00:01:34.340 to build a real functioning web app from scratch. 00:01:34.459 --> 00:01:36.299 And we're going to be really specific. We're 00:01:36.299 --> 00:01:38.640 analyzing a guide on modern app creation that 00:01:38.640 --> 00:01:41.599 uses two main tools, Notebook LM for the planning 00:01:41.599 --> 00:01:44.040 phase, and then Google Anti -Gravity for the 00:01:44.040 --> 00:01:46.159 actual building. Right. And for anyone listening 00:01:46.159 --> 00:01:47.760 who's thinking, oh, I don't need to build software, 00:01:47.900 --> 00:01:50.200 I'm a manager, just stick with us. Because this 00:01:50.200 --> 00:01:51.840 isn't really about software. It's about how we 00:01:51.840 --> 00:01:54.950 create things now. We're going to walk through 00:01:54.950 --> 00:01:58.349 building an AI invoice generator. I know it sounds 00:01:58.349 --> 00:02:01.030 boring, maybe. It sounds a little dry. But wait 00:02:01.030 --> 00:02:02.870 until you hear the numbers behind it. It's the 00:02:02.870 --> 00:02:05.290 perfect case study. It strips away all the hype 00:02:05.290 --> 00:02:08.050 and just focuses on pure utility, shows you the 00:02:08.050 --> 00:02:11.009 plumbing of the new internet. So here's our roadmap. 00:02:11.569 --> 00:02:14.090 First, we're going to talk about this huge philosophy 00:02:14.090 --> 00:02:17.580 shift. why Clarity has replaced coding, and why 00:02:17.580 --> 00:02:19.780 that's actually harder for some people. Yeah, 00:02:19.780 --> 00:02:21.620 that's a big one. Then we'll look at the brain 00:02:21.620 --> 00:02:24.599 of the operation, using Notebook LM to plan. 00:02:25.120 --> 00:02:27.960 Then the builder, using Google Anti -Gravity 00:02:27.960 --> 00:02:31.000 to execute. And finally, how to scale it so it's 00:02:31.000 --> 00:02:34.080 not just a toy, but a real product with a backend 00:02:34.080 --> 00:02:37.020 and users. It's the complete life cycle, from 00:02:37.020 --> 00:02:39.020 just a vague thought all the way to a deployed 00:02:39.020 --> 00:02:41.599 application. Let's jump into that big realization 00:02:41.599 --> 00:02:45.449 first, the why. Ah. You mentioned the bottleneck 00:02:45.449 --> 00:02:47.210 shifted. I want to dig into that, because I feel 00:02:47.210 --> 00:02:49.629 like I've tried to use AI to build stuff before. 00:02:50.009 --> 00:02:52.050 And frankly, I usually get garbage. You are not 00:02:52.050 --> 00:02:53.930 alone. That's the most common complaint I hear. 00:02:54.289 --> 00:02:57.169 AI can't code complex apps. I'll admit it. I'll 00:02:57.169 --> 00:02:59.830 go into a chat and type, build me a cool productivity 00:02:59.830 --> 00:03:02.550 app. And it spits out something that looks like 00:03:02.550 --> 00:03:04.750 it was made in 1998. The buttons don't work. 00:03:04.889 --> 00:03:07.210 And it just hallucinates stitchers. It feels 00:03:07.210 --> 00:03:09.479 like the promise is broken. That's the classic 00:03:09.479 --> 00:03:11.780 trap. See, in the past, the really hard part 00:03:11.780 --> 00:03:14.479 was the translation. You had an idea. I want 00:03:14.479 --> 00:03:16.599 a to -do list. And you had to translate that 00:03:16.599 --> 00:03:18.800 into Python or JavaScript. That was the hard 00:03:18.800 --> 00:03:22.099 labor. That was the skill. Right. Now, the AI 00:03:22.099 --> 00:03:25.159 handles the translation. And it handles it perfectly. 00:03:25.580 --> 00:03:28.419 So if the output is bad, it's not because the 00:03:28.419 --> 00:03:30.479 translator failed. It's because the source material, 00:03:30.639 --> 00:03:33.479 your instruction was bad. Wow. OK, so that's 00:03:33.479 --> 00:03:35.199 a tough pill to swallow. You're basically saying 00:03:35.199 --> 00:03:37.659 my vague prompts are just exposing my own vague 00:03:37.659 --> 00:03:41.840 thinking. I mean, yeah. In a way, exactly. If 00:03:41.840 --> 00:03:44.960 you say, build me an app, the AI has to make 00:03:44.960 --> 00:03:47.520 up all the details. It has to guess the user 00:03:47.520 --> 00:03:49.860 flow, the color scheme, the database, and it 00:03:49.860 --> 00:03:52.020 guesses based on averages. So you just get a 00:03:52.020 --> 00:03:54.419 generic mess. So the new hard skill is in syntax. 00:03:54.879 --> 00:03:57.900 The new hard skill in 2026 is product architecture. 00:03:58.419 --> 00:04:00.319 It's knowing exactly what to build and explaining 00:04:00.319 --> 00:04:03.580 it with extreme precision. So if code isn't the 00:04:03.580 --> 00:04:06.930 bottleneck anymore, what stops people now? Why 00:04:06.930 --> 00:04:08.509 isn't everyone walking around with a million 00:04:08.509 --> 00:04:10.409 dollar app in their pocket? It's the discipline 00:04:10.409 --> 00:04:12.889 of the plan. Most people want to skip right to 00:04:12.889 --> 00:04:14.689 the building part. They want that dopamine hit 00:04:14.689 --> 00:04:17.610 of seeing the screen light up. But the winners, 00:04:18.430 --> 00:04:20.430 the people actually shipping products, they're 00:04:20.430 --> 00:04:22.689 the ones who spent 80 % of their time defining 00:04:22.689 --> 00:04:25.529 the blueprint before they ever asked the AI to 00:04:25.529 --> 00:04:27.970 write a single line of code. That makes so much 00:04:27.970 --> 00:04:30.370 sense. It's like trying to build a house by telling 00:04:30.370 --> 00:04:33.149 a contractor, just build me a nice house. You're 00:04:33.149 --> 00:04:34.490 going to end up with a weird house. You'll end 00:04:34.490 --> 00:04:36.389 up with a kitchen in the bedroom and no front 00:04:36.389 --> 00:04:39.750 door. So essentially, the barrier is no longer 00:04:39.750 --> 00:04:42.389 technical capability, but the clarity of the 00:04:42.389 --> 00:04:45.769 human intent. Correct. Vague inputs create vague 00:04:45.769 --> 00:04:48.209 software. Right. OK, so let's talk about the 00:04:48.209 --> 00:04:51.110 what. The source material we're looking at suggests 00:04:51.110 --> 00:04:54.829 building an invoice generator. I have to be honest, 00:04:54.850 --> 00:04:57.839 when I think of revolutionary tech, My mind does 00:04:57.839 --> 00:05:01.360 not jump to invoice generator. That sounds incredibly 00:05:01.360 --> 00:05:05.139 dry. Why this? It comes down to what we call 00:05:05.139 --> 00:05:07.519 boring scale. Yeah. You can chase the flashy, 00:05:07.800 --> 00:05:09.959 never -before -seen ideas, you know, Uber for 00:05:09.959 --> 00:05:12.800 dog walkers on Mars, but those are so high risk, 00:05:12.839 --> 00:05:15.540 they require changing user behavior. An invoice 00:05:15.540 --> 00:05:19.259 generator. It addresses a constant annoying problem. 00:05:19.680 --> 00:05:21.860 Every freelancer, every small business, every 00:05:21.860 --> 00:05:24.220 contractor needs to send invoices. And the numbers 00:05:24.220 --> 00:05:27.870 in the source material were... Yeah. I had no 00:05:27.870 --> 00:05:29.850 idea how much traffic these basic tools get. 00:05:30.069 --> 00:05:33.009 They're huge. Some of the basic invoice tools 00:05:33.009 --> 00:05:35.170 out there, tools that look like they have been 00:05:35.170 --> 00:05:38.910 updated in a decade, are getting over 2 .1 million 00:05:38.910 --> 00:05:43.310 visits a month. Wait, 2 .1 million a month for 00:05:43.310 --> 00:05:45.829 a simple form that just spits out a PDF? That's 00:05:45.829 --> 00:05:48.310 right. Because it solves an immediate pain. And 00:05:48.310 --> 00:05:50.529 the source makes a great point here. You don't 00:05:50.529 --> 00:05:52.209 need to reinvent the wheel. You just need to 00:05:52.209 --> 00:05:54.449 be slightly better. Right. If you can build a 00:05:54.449 --> 00:05:56.649 tool that's cleaner, faster, or has better AI 00:05:56.649 --> 00:05:59.350 defaults than the old ones, you only need a tiny 00:05:59.350 --> 00:06:01.850 fraction of that 2 .1 million to have a very 00:06:01.850 --> 00:06:03.769 successful product. It's like the unbundling 00:06:03.769 --> 00:06:05.629 of software. You don't need to be Salesforce. 00:06:05.790 --> 00:06:08.709 You just need to be the best tiny tool for one 00:06:08.709 --> 00:06:11.850 specific person. Exactly. And Google Trends shows 00:06:11.850 --> 00:06:14.949 the search for AI invoice generator is just a 00:06:14.949 --> 00:06:16.629 vertical line going up. The demand is there. 00:06:16.629 --> 00:06:19.470 You just have to meet it. So does a builder need 00:06:19.470 --> 00:06:22.329 a revolutionary, never -before -seen idea to 00:06:22.329 --> 00:06:25.430 succeed? No, just improve an existing solution 00:06:25.430 --> 00:06:27.509 that already has high demand. Okay, so we have 00:06:27.509 --> 00:06:29.670 our idea. Invoice generator. We know the demands 00:06:29.670 --> 00:06:33.769 there. Now, step one. The source calls this the 00:06:33.769 --> 00:06:36.990 brain. This is where Notebook LM comes in. Why 00:06:36.990 --> 00:06:38.930 aren't we going straight to the builder? Why 00:06:38.930 --> 00:06:41.069 not just open up a chat and say start coding? 00:06:41.689 --> 00:06:43.250 Because of what we just talked about. If you 00:06:43.250 --> 00:06:45.089 go straight to the builder, you'll be vague. 00:06:45.629 --> 00:06:48.329 Notebook LM is like your architect. It forces 00:06:48.329 --> 00:06:50.410 you to slow down. Walk me through this workflow. 00:06:50.589 --> 00:06:53.519 I go to notebooklm .google. Then what? You don't 00:06:53.519 --> 00:06:56.560 just dump a random sentence in. The guide suggests 00:06:56.560 --> 00:06:59.100 using a prompt builder first. It's basically 00:06:59.100 --> 00:07:01.620 a template to define your constraints. You define 00:07:01.620 --> 00:07:04.220 the niche, which is invoice generator, your skill 00:07:04.220 --> 00:07:07.420 level, say, beginner, and the style, like minimal 00:07:07.420 --> 00:07:10.439 or modern. You paste that structured idea into 00:07:10.439 --> 00:07:12.920 notebook LM. And what does notebook LM do with 00:07:12.920 --> 00:07:15.139 that? Isn't it just like a document reader? You'd 00:07:15.139 --> 00:07:17.540 think so, but it's... It's more than that. It 00:07:17.540 --> 00:07:19.540 sort of acts like a mirror. It processes that 00:07:19.540 --> 00:07:21.879 intent, and then it interviews you. It summarizes 00:07:21.879 --> 00:07:24.240 your idea back to you. It suggests features you 00:07:24.240 --> 00:07:27.199 hadn't thought of. It outlines a user flow. And 00:07:27.199 --> 00:07:30.220 it asks these critical questions like, how will 00:07:30.220 --> 00:07:33.040 you handle currency conversion? That's incredibly 00:07:33.040 --> 00:07:34.819 useful. It's like having a product manager sitting 00:07:34.819 --> 00:07:37.040 next to you asking, OK, but what happens when 00:07:37.040 --> 00:07:39.980 they click Save? Exactly. It finds the holes 00:07:39.980 --> 00:07:42.000 in your logic before you build them. And the 00:07:42.000 --> 00:07:43.899 crucial output of this step, the whole point 00:07:43.899 --> 00:07:46.980 of it, is what the source calls the final prompt. 00:07:47.519 --> 00:07:49.600 Define that for me. What does a final prompt 00:07:49.600 --> 00:07:51.980 look like? Is it just a sentence? No, no, it's 00:07:51.980 --> 00:07:54.420 not a sentence. It's a comprehensive text file. 00:07:54.600 --> 00:07:57.300 Sometimes it's two or three pages long. It details 00:07:57.300 --> 00:07:59.779 the color palette, the typography, the button 00:07:59.779 --> 00:08:02.939 behaviors, the database schema, even the error 00:08:02.939 --> 00:08:06.079 messages. It is a precise set of instructions 00:08:06.079 --> 00:08:09.860 that leaves zero room for ambiguity. So the physical 00:08:09.860 --> 00:08:13.500 output of this entire first step is... What exactly? 00:08:13.680 --> 00:08:16.319 A single highly detailed prompt that acts as 00:08:16.319 --> 00:08:18.699 the app's blueprint. Okay, I have my blueprint. 00:08:18.819 --> 00:08:20.959 I have my final prompt. Now I'm ready to actually 00:08:20.959 --> 00:08:23.279 build. This is where we switch tools. We're moving 00:08:23.279 --> 00:08:26.759 to the builder, which is Google Anti -Gravity. 00:08:27.040 --> 00:08:28.519 Love the name, by the way. It is pretty cool. 00:08:28.899 --> 00:08:32.139 So Anti -Gravity is the execution engine. And 00:08:32.139 --> 00:08:33.860 unlike the web chats we're used to, this is a 00:08:33.860 --> 00:08:36.100 desktop app. You open it up and it looks like 00:08:36.100 --> 00:08:38.820 a chat interface, but it has permissions to access 00:08:38.820 --> 00:08:41.190 your local files. Why does that matter? Why on 00:08:41.190 --> 00:08:43.429 my desktop? Because it's not just giving you 00:08:43.429 --> 00:08:45.909 code snippets to copy and paste. It's actually 00:08:45.909 --> 00:08:48.850 creating files on your hard drive. You take that 00:08:48.850 --> 00:08:51.330 final prompt from Notebook LM, you paste it in, 00:08:51.549 --> 00:08:54.009 and this is a key detail from the source. You 00:08:54.009 --> 00:08:57.159 select the Claude Opus model. Why that specific 00:08:57.159 --> 00:09:00.179 model? Why not Gemini or GPT -4? In the context 00:09:00.179 --> 00:09:03.039 of these 2026 tools, the source really highlights 00:09:03.039 --> 00:09:05.120 Claude Opus for having the highest reliability 00:09:05.120 --> 00:09:08.240 for complex coding tasks. It creates fewer bugs, 00:09:08.460 --> 00:09:10.919 it structures the code better, and it hallucinates 00:09:10.919 --> 00:09:13.720 less. Might be a bit slower, but for coding, 00:09:14.179 --> 00:09:17.799 accuracy is everything. OK, so I paste the blueprint, 00:09:17.940 --> 00:09:21.000 I select Opus, I hit enter. What happens? Describe 00:09:21.000 --> 00:09:23.600 the experience for me. The dev team in a box 00:09:23.600 --> 00:09:26.549 wakes up. It reads the blueprint and immediately 00:09:26.549 --> 00:09:30.549 starts creating files. It writes the index .html 00:09:30.549 --> 00:09:33.669 for the structure, the style .css for the design, 00:09:33.730 --> 00:09:36.529 the script .js for the logic. It does it all 00:09:36.529 --> 00:09:38.529 at once. And the timeline for that. The source 00:09:38.529 --> 00:09:41.269 says about five to 10 minutes. That is wild. 00:09:41.649 --> 00:09:43.590 I mean, five to 10 minutes to do what would have 00:09:43.590 --> 00:09:46.149 taken a junior developer weeks. And you aren't 00:09:46.149 --> 00:09:47.990 managing it. You're just watching it happen. 00:09:48.049 --> 00:09:50.110 You see the files just appear in your folder 00:09:50.110 --> 00:09:52.370 in real time. It's almost spooky. OK, here's 00:09:52.370 --> 00:09:55.549 where I get skeptical. Because nothing is ever 00:09:55.549 --> 00:09:57.610 perfect on the first try, right? Even with a 00:09:57.610 --> 00:10:00.309 great prompt. Let's say the app pops out, I open 00:10:00.309 --> 00:10:04.049 it in my browser, and the green is ugly. Or a 00:10:04.049 --> 00:10:06.190 button is in the wrong spot. I don't know, CSS. 00:10:06.350 --> 00:10:08.750 Am I just stuck? Not at all. And this is the 00:10:08.750 --> 00:10:11.429 beauty of this zero code workflow. You don't 00:10:11.429 --> 00:10:13.730 open the code file. You go back to the chat window 00:10:13.730 --> 00:10:16.279 in anti -gravity. You treat it like a designer 00:10:16.279 --> 00:10:18.299 sitting right there with you. You just say, make 00:10:18.299 --> 00:10:21.460 the layout simpler or change that button to dark 00:10:21.460 --> 00:10:23.419 green. So wait, I'm not digging into code. I'm 00:10:23.419 --> 00:10:25.960 just giving feedback in plain English. It's just 00:10:25.960 --> 00:10:29.080 plain English. And because antigravity has access 00:10:29.080 --> 00:10:31.340 to the files it just wrote, It knows exactly 00:10:31.340 --> 00:10:33.759 what to change. It updates the files for you 00:10:33.759 --> 00:10:36.500 instantly. What about actual bugs? Like, if I 00:10:36.500 --> 00:10:38.919 click generate and nothing happens? You just 00:10:38.919 --> 00:10:40.919 describe the error. Hey, when I click generate, 00:10:41.019 --> 00:10:43.519 nothing happens. The AI investigates its own 00:10:43.519 --> 00:10:46.279 code, finds the missing logic, maybe a broken 00:10:46.279 --> 00:10:48.559 event listener or something, and it fixes it. 00:10:48.919 --> 00:10:52.139 So how does a non -coder fix a broken feature? 00:10:52.460 --> 00:10:54.620 You just describe the error in plain English 00:10:54.620 --> 00:10:57.179 to the chat. Okay, so now we have a front end. 00:10:57.399 --> 00:10:59.820 It looks good, it works, I can generate a PDF, 00:11:00.279 --> 00:11:02.759 but the source mentions a level up step. The 00:11:02.759 --> 00:11:04.820 backend. Right, because initially what you've 00:11:04.820 --> 00:11:07.080 built is a static tool. It just lives in your 00:11:07.080 --> 00:11:09.419 browser. If you refresh the page, all your data 00:11:09.419 --> 00:11:12.159 is gone. It's what developers call stateless. 00:11:12.360 --> 00:11:13.980 Which is fine for a one -off thing, but if I 00:11:13.980 --> 00:11:15.980 want to run a business, I want to save my invoices, 00:11:16.100 --> 00:11:18.840 I want a history, I want customer profiles. Exactly. 00:11:19.580 --> 00:11:21.659 And traditionally, this is where the difficulty 00:11:21.659 --> 00:11:24.360 curve goes vertical. Add a backend usually means 00:11:24.360 --> 00:11:26.759 setting up a server, configuring a database, 00:11:27.299 --> 00:11:31.279 handling security, API keys. It's a nightmare 00:11:31.279 --> 00:11:33.980 for beginners. Ugh, I'm getting a headache just 00:11:33.980 --> 00:11:36.700 listening to that list. Yeah. That's the part 00:11:36.700 --> 00:11:38.279 where everyone quits. That's always the wall 00:11:38.279 --> 00:11:40.980 people hit. But in this anti -gravity workflow, 00:11:41.179 --> 00:11:43.879 the prompt is just absurdly simple. You just 00:11:43.879 --> 00:11:46.600 tell the chat, add a sign -up and login system 00:11:46.600 --> 00:11:49.440 so users can save invoices. That's it. You don't 00:11:49.440 --> 00:11:51.379 have to specify what kind of database or where 00:11:51.379 --> 00:11:54.220 to host it. Nope. The AI understands that the 00:11:54.220 --> 00:11:57.080 request save invoices implies a database and 00:11:57.080 --> 00:12:00.039 authentication. The source notes that it'll likely 00:12:00.039 --> 00:12:02.679 integrate something like Firebase or SuperBase 00:12:02.679 --> 00:12:05.840 automatically. It sets up the auth rules, creates 00:12:05.840 --> 00:12:07.919 the database structure, and connects everything. 00:12:08.139 --> 00:12:10.539 Does adding a database require me to start a 00:12:10.539 --> 00:12:13.659 new project file? No, you just continue the same 00:12:13.659 --> 00:12:17.139 chat and the AI evolves the existing app. That 00:12:17.139 --> 00:12:19.460 is just incredibly efficient. So now we have 00:12:19.460 --> 00:12:21.860 a working app. It has a login system. But an 00:12:21.860 --> 00:12:24.059 app without users is just a file on my computer. 00:12:24.879 --> 00:12:26.919 We need to talk about marketing. The part everyone 00:12:26.919 --> 00:12:29.340 fears, the build it and they will come fallacy. 00:12:29.539 --> 00:12:32.100 Right. And the source suggests going back to 00:12:32.100 --> 00:12:35.320 Notebook LM for this. Why switch back? Because 00:12:35.320 --> 00:12:38.740 antigravity is for building. Notebook LM is for 00:12:38.740 --> 00:12:41.639 thinking. You go back to the studio section in 00:12:41.639 --> 00:12:43.419 Nopagalem, where it already understands your 00:12:43.419 --> 00:12:46.259 project, and you prompt it. Give me a step -by 00:12:46.259 --> 00:12:49.320 -step plan to get traffic to my AI invoice generator. 00:12:49.720 --> 00:12:52.059 And because it knows the niche, it's not giving 00:12:52.059 --> 00:12:54.840 you generic advice like post on Instagram. Exactly. 00:12:54.980 --> 00:12:56.440 It's not going to tell you to start a TikTok 00:12:56.440 --> 00:12:59.059 dance if that doesn't fit. The strategy the source 00:12:59.059 --> 00:13:02.200 highlights is SEO and content. It suggests creating 00:13:02.200 --> 00:13:04.440 content around terms people are already searching 00:13:04.440 --> 00:13:07.440 for, like free invoice generator template or 00:13:07.440 --> 00:13:09.960 AI invoice generator for small business. And 00:13:09.960 --> 00:13:12.600 the app itself becomes the solution inside that 00:13:12.600 --> 00:13:14.980 content. Precisely. You're not aggressively selling. 00:13:15.220 --> 00:13:18.179 You're providing value. And the tool is the payoff 00:13:18.179 --> 00:13:20.840 right there. So what's the primary goal of this 00:13:20.840 --> 00:13:23.019 first version then? It's not perfection, is it? 00:13:23.159 --> 00:13:25.860 Not at all. The goal is just feedback. You want 00:13:25.860 --> 00:13:28.120 to get it into the hands of real users, even 00:13:28.120 --> 00:13:30.820 just 10 people, so they can tell you what's missing 00:13:30.820 --> 00:13:33.460 or what's broken. That feedback loop is what 00:13:33.460 --> 00:13:36.299 turns a project into a product. It really seems 00:13:36.299 --> 00:13:38.879 like the biggest risk here isn't technical failure 00:13:38.879 --> 00:13:41.480 anymore. It's just not doing it. The cost of 00:13:41.480 --> 00:13:43.940 failure has dropped to zero. The tools are free 00:13:43.940 --> 00:13:47.000 or cheap. The time investment is hours, not months. 00:13:47.039 --> 00:13:49.659 The only cost left is your own hesitation. So 00:13:49.659 --> 00:13:52.139 the primary goal of version 1 is simply user 00:13:52.139 --> 00:13:55.429 feedback. Exactly. Not perfection, but feedback 00:13:55.429 --> 00:13:59.029 from real users to guide improvements. We're 00:13:59.029 --> 00:14:00.629 going to take a quick break, but when we come 00:14:00.629 --> 00:14:03.350 back, I want to recap this Lego block philosophy 00:14:03.350 --> 00:14:05.490 and give you one final thought that might just 00:14:05.490 --> 00:14:08.029 change how you look at every problem you encounter 00:14:08.029 --> 00:14:13.690 today. Stay with us. And we are back on the deep 00:14:13.690 --> 00:14:16.929 dive. We've been unpacking this zero code builder 00:14:16.929 --> 00:14:20.789 workflow using Notebook LM to plan and Google 00:14:20.789 --> 00:14:23.299 anti -gravity to build. It's really a fundamental 00:14:23.299 --> 00:14:26.000 shift, isn't it? It democratizes the actual act 00:14:26.000 --> 00:14:28.480 of building something. It really feels like playing 00:14:28.480 --> 00:14:30.320 with Lego blocks. You just have these distinct 00:14:30.320 --> 00:14:33.139 pieces that snap together. That's a perfect analogy. 00:14:33.299 --> 00:14:35.539 Think about it. You have the idea block. You 00:14:35.539 --> 00:14:37.720 combine that with the notebook LM block to get 00:14:37.720 --> 00:14:40.200 a plan. Then you snap that plan into the anti 00:14:40.200 --> 00:14:42.139 -gravity block to get an app. And then you add 00:14:42.139 --> 00:14:44.279 the user feedback block to get a real product. 00:14:44.620 --> 00:14:47.059 And none of those blocks require you to know 00:14:47.059 --> 00:14:49.659 how to injection mold plastic to keep the metaphor 00:14:49.659 --> 00:14:51.720 going. You just have to know how to stack them. 00:14:51.950 --> 00:14:54.830 Right. You're the architect, not the bricklayer. 00:14:55.230 --> 00:14:57.470 And honestly, it's so exciting because it removes 00:14:57.470 --> 00:15:01.090 that whole intimidation factor. I think the thing 00:15:01.090 --> 00:15:03.049 that sticks with me most from this deep dive 00:15:03.049 --> 00:15:06.070 is the idea that waiting is now the only real 00:15:06.070 --> 00:15:09.470 mistake. It is. In 2026, if you have an idea 00:15:09.470 --> 00:15:11.350 and you don't build it, it's not because you 00:15:11.350 --> 00:15:13.289 couldn't, it's because you didn't start. That 00:15:13.289 --> 00:15:15.269 old excuse, I don't have a technical co -founder 00:15:15.269 --> 00:15:18.549 or I don't have the budget is gone. That's a 00:15:18.549 --> 00:15:20.519 little haunting. but also really empowering. 00:15:21.080 --> 00:15:23.340 It puts the responsibility squarely back on us. 00:15:23.840 --> 00:15:26.240 It forces a shift in mindset. We are moving from 00:15:26.240 --> 00:15:29.259 a world of how do I build this, which is a technical 00:15:29.259 --> 00:15:31.919 question, to what deserves to be built, which 00:15:31.919 --> 00:15:34.279 is a very human question. What deserves to be 00:15:34.279 --> 00:15:37.080 built? That is the question. Because now we can 00:15:37.080 --> 00:15:39.860 build almost anything. The bottleneck is no longer 00:15:39.860 --> 00:15:43.000 the machine. It's our own imagination and our 00:15:43.000 --> 00:15:45.779 ability to clarify our thoughts. So here's our 00:15:45.779 --> 00:15:47.840 challenge to you listening right now. You probably 00:15:47.840 --> 00:15:51.399 have that one idea. that little tool you wish 00:15:51.399 --> 00:15:54.340 existed for your work, or that hobby project 00:15:54.340 --> 00:15:57.259 you've always put off. Open Notebook LM. Just 00:15:57.259 --> 00:15:59.879 do the first step. Exactly. Don't worry about 00:15:59.879 --> 00:16:02.320 the database or the marketing yet. Just go to 00:16:02.320 --> 00:16:05.960 Notebook LM, type in your idea, and let it generate 00:16:05.960 --> 00:16:08.360 a plan for you. Just see what the blueprint looks 00:16:08.360 --> 00:16:10.980 like. You might be surprised at how real it feels 00:16:10.980 --> 00:16:13.139 once you see it all written down. It stops being 00:16:13.139 --> 00:16:16.039 a dream and starts being a to -do list. And once 00:16:16.039 --> 00:16:18.299 it's a to -do list... where you're only about 00:16:18.299 --> 00:16:20.139 10 minutes away from having it running on your 00:16:20.139 --> 00:16:22.700 computer. And that is a very exciting place to 00:16:22.700 --> 00:16:24.820 be. Thank you so much for joining us on this 00:16:24.820 --> 00:16:27.159 deep dive. It's been a fascinating look at the 00:16:27.159 --> 00:16:29.120 future of building. I think I'm going to go try 00:16:29.120 --> 00:16:31.559 to build that productivity app again, but this 00:16:31.559 --> 00:16:34.299 time I promise to be specific. Let me know how 00:16:34.299 --> 00:16:36.879 the blueprint turns out. Will do. Thanks for 00:16:36.879 --> 00:16:37.860 listening, everyone. Take care.