Jacob Klug
@Jacobsklug • 32,934 subscribers
Building AI-native agencies. CEO @ https://t.co/z3PEeohxf2
Shorts
Videos
I've made $4.7M with AI. I'm giving away the exact prompts I used to build a full AI agency from scratch in under 2 hours. These prompts will guide you through: • Finding a winning niche • Building an offer people pay for • Go to market strategy • Creating your website • Productizing with AI (ie Lovable) This took me 5 years and hundreds of thousands of dollars to figure out. Comment "Build" and follow. I'll DM it to you. P.S. This will probably blow up so give me some time to reply.
Jacob Klug137,904 次观看 • 2 个月前
I've made $4.7M with AI. Today, I recorded myself building an entire OpenClaw🦞 business in an hour. You'll learn how to: • Find a winning niche using AI • Build an offer people actually want to pay for • To get your first clients • To use AI to productize and fulfill the entire service • The exact prompts I use to orchestrate all of it After building the #1 Lovable agency. This is the entire playbook behind AI native businesses. The full video is live on YouTube. Comment "Playbook" and I'll DM you the link.
Jacob Klug123,701 次观看 • 2 个月前
These vibe-coded apps make $180K/month. With the smallest making $3k/month all the way to $90k/month. In this video, I go through each and analyzed what make each work (or not work). So you can replicate these strategies in your own apps. Comment and I'll DM you the full video.
Jacob Klug113,336 次观看 • 5 个月前
This design system made us $2.5M. I'm giving it away for free to Lovable builders. This will allow you to build premium UI in Lovable. Copy & paste. What’s included: • 3 design styles • Component standards (buttons, cards, nav, modals) • Copy right into project Follow + comment "Design" and I'll DM you.
Jacob Klug74,082 次观看 • 4 个月前
I tried to clone a $50M/year SaaS product using only AI in under 30 minutes. To see if I could build & run it 100x cheaper. The first product that came to mind was Trello—used by millions, charges $10/user/month, and at its core, it's a simple kanban board. The result: a fully functional Trello clone built in ~23 minutes with Lovable. Saving users millions of dollars. Drag and drop, team invites, real-time sync, comments, multiple boards. Here's the technique and full breakdown:
Jacob Klug62,035 次观看 • 4 个月前
This is how I built a OpenClaw🦞 AI operating system that runs my entire life and business. Here's what I cover ↓ 0:00 - The Operating System Overview 0:42 - Setting Up Your Clawdbot's Identity (Soul + User Files) 1:28 - The Memory System That Actually Works 3:12 - Organizing Meetings With Fathom Integration 4:05 - Giving Your Team Access Through Slack 4:38 - Automated Cron Jobs (Daily Briefs, Meeting Syncs, Content Drafts) 6:18 - Why I Stopped Building Dashboards in OpenClaw 6:52 - Scoping the OS With Your Clawdbot 8:30 - Building the Interface in Lovable (Live) 10:15 - Connecting Supabase + Setting Up the Database 11:40 - REST APIs So Your Bot Controls the Dashboard 12:45 - The Final Product: Tasks, Projects, Content, All in One Place Bookmark for later & follow me for more builds like this.
Jacob Klug47,446 次观看 • 3 个月前
Everyone is hyping Paperclip as the OpenClaw🦞 killer. I spent a day building an entire AI company on it. Here's my honest take. In this video you'll see: • How Paperclip spins up an org chart, CEO, and sub agents in under 5 minutes • The exact prompt I used to bootstrap a content agency from scratch • Where it actually beats Open Claw (and where it falls flat) • Why I think it's the iPod of agent orchestration, not the iPhone • The 12 page architecture doc I built alongside it Full video is live on YouTube. Comment "Paperclip" and I'll DM you the link.
Jacob Klug23,407 次观看 • 1 个月前
Everyone is talking about OpenClaw mission controls. I rebuilt a $250K mission control using only one prompt. In this video you'll see: •The 3 levels of mission controls real builders are running •Why a mission control is critical to OpenClaw success •The exact prompt I used to generate the full dashboard in Lovable •My agent creating tasks in real time on the board instantly The full video is live on YouTube. Comment "Mission" and I'll DM you the link.
Jacob Klug27,911 次观看 • 3 个月前
I now have my own AI on my mac. I just found the upgraded Openclaw without the 6 hour setup time. That can talk, complete tasks, respond to emails & more. I don't need to even touch my keyboard anymore. It took me ~20 minutes to build this AI with Lemon. So far been quite impressed - worth giving a shot.
Jacob Klug29,209 次观看 • 3 个月前
I ranked the 3 best OpenClaw🦞 mission controls I've seen — then rebuilt the best one live. Here's what I cover ↓ 0:00 - The 3 Mission Controls That Can Power Your OpenClaw 0:42 - Sub-Agent Management (Clean, Simple, Effective) 2:15 - The Task-Focused Dashboard Going Semi-Viral 4:05 - Live Activity Feeds, Memory, Docs & Chat In One Place 5:30 - Alex Finn's Full Build (The King of OpenClaw) 7:00 - Content Pipelines, Cron Job Calendars & Daily Journals 8:30 - Why I Rebuild These in Lovable Instead of OpenClaw 9:15 - Using OpenClaw to craft Lovable Prompt 10:00 - The One-Prompt Dashboard Build (Live) 11:00 - The Final Product — Tasks, Content, Calendar, Memory, Team & Settings Bookmark this for when you're ready to build yours. Follow for more builds like this.
Jacob Klug22,865 次观看 • 2 个月前
We just vibe-coded a tool for a law firm. Built entirely with AI. No engineers. No code. It uses a custom model (trained on their SOPs) to review legal documents and auto-save them to the right case file. It’s saving them thousands each month—in software fees and hours of manual work.
Jacob Klug64,407 次观看 • 1 年前
I'm so tired of designers thinking AI won't replace them. What we watched with vs Brett wasn’t a win for designers. Let’s take a moment to think about this. A year ago, the idea of an AI going head-to-head with a professional designer would’ve been laughable. Then it happened. And the odds were real. Brett the DJ said it himself — he was shaking on the call. He won, but it wasn’t a blowout by any means. Then he went on a victory parade, talking about how AI will never reach his god-like levels. But let’s be clear: We’re in the iPhone 1 days right now. AI is going to get exponentially better. And Brett? He’s already stagnated — if not declining — as he inches closer to retirement. What we’ll likely see (and are already seeing) is the merge of designer and engineer into one role: The design engineer. > Someone with taste. > Knows how to prompt. > Has baseline technical skills. > And can build/design directly in AI-native tools. Figma should be worried. And I think they know it. The traditional “design phase” is becoming obsolete. Why mock things up when you can run live experiments faster in dev?
Jacob Klug64,921 次观看 • 1 年前
This is INSANE. My OpenClaw🦞 just called booked me a reservation. Here’s what actually happened: 1/ The Setup I gave my OpenClaw a real phone number. Most people would say that’s risky. I think the opposite. Why sit on hold when AI can do it for you? I installed the Clawdtalk skill, generated a phone number + PIN, and was live in minutes. That’s it. No complex setup. 2/ The Test Valentine’s Day was coming up. I told OpenClaw: “Find me a restaurants near me for 2 people on Saturday.” I genuinely didn’t think it would work. It called 5 restaurants. At the same time. Spoke to real humans. Booked 1 with availability. No hold music. No back and forth. No wasted time. 3/ The Shift The infrastructure is changing. OpenClaw isn’t just a chatbot. It’s an execution layer. The interface to the real world is becoming API-accessible. Restaurants. Doctors. Contractors. Front desks. If it has a phone number, it’s now programmable. Today: dinner reservations. Tomorrow: vendor negotiations, recruiting screens, appointment backlogs. Anything that requires: – calling – waiting – confirming – coordinating …can now be delegated. We’re entering the era of agentic execution. I spoke with the Clawdtalk team and got a free usage tier for builders in my audience. Comment “Clawdtalk” and I’ll DM you the link. Let’s build.
Jacob Klug18,864 次观看 • 3 个月前
I built this deck in ~15 minutes using Lovable. Clickable, animated, shareable. Analytics built in natively. A few years ago, this would’ve taken days (if not weeks). Today, a few prompts handle the slides, visuals, copy, and tracking. PS. Anton Osika invest in my startup? 😅
Jacob Klug21,170 次观看 • 4 个月前
6 months ago, building an app required: - 6 months of development - $300K budget - Team of 5 developers - Constant manual debugging Today, I'm helping a 17-year-old build one in a week. Workflow: ChatGPT -> Lovable -> Supabase -> Cursor Step-by-step how: 1/ Write your PRD first. This is non-negotiable. Clear requirements on paper = 10x faster execution with AI. We spent 30 minutes documenting exactly what PostPal needed to do. Then gave it to ChatGPT: "Create a Lovable prompt from this PRD." Copy. Paste. Done. 2/ Your first prompt sets everything. That initial Lovable prompt? It's your foundation. Give it the full high-level vision. Every feature. Every flow. Lovable uses this context for everything that follows. Get this wrong and you'll rebuild from scratch. 3/ Database first, frontend second. Biggest mistake I see: Building the entire UI, then trying to connect data. Set up Supabase immediately. Create your tables. Configure role-level security. Build backend and frontend together, not separately. 4/ Go feature by feature. Don't attack all screens at once. Pick one core feature. Build it completely. Connect it to your database. Test it. Then move to the next. Each feature should be fully functional before moving on. 5/ Chat mode is your debugging superpower. When something breaks (it will): → Use chat mode to diagnose → Let it explain the issue → Switch to agent mode to fix Chat mode for understanding. Agent mode for implementing. This combo saved us hours. 6/ Security isn't optional. Before deploying: → Enable row-level security → Secure your edge functions → Check your API endpoints → Run Lovable's security check Takes 5 minutes. Saves you from disasters. The result? A fully functional app. Not a prototype. Not a demo. A real product with authentication, database, and payments. The game has completely changed. While others debate if AI will replace developers... We're shipping products before lunch. The tools are here. The playbook is proven. The only question is: what will you build? Time to ship.
Jacob Klug40,338 次观看 • 10 个月前