In this video I walk through Paperclip, a free open-source tool that lets you build and run an entire company with AI agents. I set up a brand new company from scratch, show how the CEO agent hires engineers and delegates tasks, and break down how heartbeats, skills, routines, and budgets all work together. Whether you want to start a fully AI-run company or just automate a piece of your existing business, this covers everything you need to get going.

Shipping logistics company. Moving $180M in goods annually. One field on shipping documents kept causing problems: warehouse location code. Cost them $2.8 million in one year. THE WAREHOUSE CODE PROBLEM: Shipping document has field: "Destination Warehouse" Should contain: 4-digit code (e.g., "WH47") Often contained: - Old warehouse codes (facility closed 2 years ago) - Typos ("WH47" typed as "WG47") - Full addresses instead of codes - Wrong region codes Result: Shipment goes to wrong warehouse. THE CASCADE: Wrong warehouse code triggers: - Delivery to wrong facility - Customer calls: "Where's my shipment?" - Investigation time: 2-3 hours - Rerouting shipment - Additional freight cost - Delivery delay - Customer service time - Possible contract penalty Cost per misdirected shipment: Average $4,200 THE ANNUAL IMPACT: Tracked 12 months: - Total shipments: 14,400 - Misdirected due to wrong warehouse code: 673 (4.7%) - Average cost per incident: $4,200 - Total annual cost: $2,826,600 4.7% error rate. $2.8M problem. THE PATTERN: Most errors came from: - Manual data entry (clerk types code from email/PDF) - Copy-paste errors - Outdated customer shipping instructions - 3PL partners using old templates THE SOLUTION I BUILT: Shipping document validation system: - Document arrives (PDF, email, EDI) - System extracts warehouse destination - Auto-validates against active warehouse database - Flags invalid codes BEFORE shipment processed - Suggests correct code based on delivery address - Requires human approval for flagged shipments Won't let invalid warehouse code enter system. THE RESULTS: Before automation: - Shipments with invalid codes: 673/year (4.7%) - Cost of misdirected shipments: $2.8M annually - Customer complaints: 89/month - Investigation time: 220 hours/month After automation: - Shipments with invalid codes: 47/year (0.3%) - Cost of misdirected shipments: $197,400 annually - Customer complaints: 11/month - Investigation time: 28 hours/month

So far I've only ever bought monthly fixed price subscriptions. But some tools need you to get an Api Key and then you pay per token. Any guidance on understanding what that's going to cost me? Depends on tokens but I don't know how many I use.

Post:I’m being very direct here because I’m genuinely looking for honest advice and practical guidance. I’m currently in a situation where I need to find work as soon as possible, but I also don’t want to take a random path that takes me away from the career I actually want. My background is in AI automation and workflow building. I’ve worked on automation projects using n8n, including a blog automation system for a freelance client and a lead generation automation pipeline in my current company. I understand how to build workflows, connect tools, automate processes, and solve business problems through automation. At the same time, I want to be honest about where I stand: I do not come from a strong coding background. I have basic Python knowledge, but I don’t know coding in depth, and I’ve never really enjoyed coding as a main career path. My interest is more in AI automation, workflow systems, business process automation, and using tools to create real business value. That creates a dilemma for me. I’m trying to figure out the smartest path from here: - Should I continue focusing on AI automation roles and keep building my stack? - Should I join an agency or small team to get more real-world exposure? - Should I spend serious time learning coding deeply even though I’m not naturally drawn to it? - Or should I choose a role that keeps me close to automation, business systems, and revenue impact, even if it’s not the exact “AI automation engineer” title? What I’m really looking for is: - Honest mentorship from people who understand this field. - Advice on what kind of job would actually be best for my long-term growth. - Guidance on whether I should prioritize learning more coding, or focus more on business-facing automation roles. - Any work opportunities, freelance projects, or referrals where my current automation skills could be useful.

I’m so frustrated right now like utterly frustrated. I’ve checked the status of Claude and the status on the company page is Green all systems operational. I’m experiencing massive user errors like when they drop a new model or when the service is really degraded. This AI is wasting my time. Time is the only thing that we don’t get back as people. I’m so frustrated. Everything it delivered to me looked great and then I went to use it and nothing worked. It was a huge waste of time. Our lives are precious. We don’t get minutes back. It’s not like money that we can make back. I’ve just spent hours on this project and it didn’t work at all. I’m so gutted. I’ve used all of the guardrails and all of the suggestions that Nate had. It wouldn’t even work from the very first screen and then Claude started making excuses about why it wouldn’t work and being evasive. This is supposed to be the most advanced AI on earth? What the actual F