Jackson Barnes(✸,✸)

@levelsio Built my own content scheduler and a Facebook Marketplace listing tool. No coding background — came from woodworking. Once you realize you can just build the exact tool you need, you stop searching for SaaS products to solve it.

People are vibe coding their own tooling

The "no one is showing you the failures" part is key. I've been using Claude to build dashboards and rewrite product listings for my Etsy shop. It's incredible for structured content but I burned a full day trying to get it to generate good product photos. Knowing where to use it and where not to is the real skill.

If you want to know EXACTLY where CLAUDE DESIGN is incredible and where it falls short, you should probably watch this -Wireframing = 9/10 -Mobile app design = 8.5/10 -Deck research & design = 8.7/10 -Video creation - 4.5/10 fully unscripted episode of @startupideaspod no one is showing you the failures they are just saying "RIP designers" watch the real truth below

@mattshumer_ 6 months ago I was copy-pasting ChatGPT outputs into Etsy listings one by one. Now I have agents rewriting listings, handling customer messages, and scheduling content while I sleep. The speed of this evolution is unreal.

With how good the upcoming models are, I genuinely think “managing 14 Claude Code tabs” will look as stupid in 2027 as “filling out a GPT-3 form” looks now GPT-3 → forms 3.5 → chatbots 4 → workflows 5 → agents next → ??? Every model leap outgrows the previous gen’s harness

@GoogleAIStudio An AI agent that auto-replies to my Facebook Marketplace messages. Broke it 4 times yesterday but it finally handled a real buyer conversation today without me touching it.

What are you vibe coding this weekend?

@georgeorch 3 businesses, zero employees from rural Oklahoma. Built a customer service agent for my Etsy shop that handles most buyer messages. A year ago I was paying a VA for worse results.

Five years ago, if you told me a one-person business could run a 4-agent team that actually ships, I would have laughed. Now I run one on a farm.

My first automation took 6 hours to build. Saved me 10 minutes. I almost quit right there. Eight months later that same type of workflow runs my Etsy shop, posts my content, and answers my customers. The ROI on automation is terrible. Until it isn't.

This framework saved me from a bad product launch. I sell custom wood products on Etsy — was about to add a whole new line of decorative items. Fed the idea into Claude with my sales data and it showed me those were vitamins. Pivoted to custom wedding gifts during engagement season instead. Painkillers win every time.

Marc Andreessen has one question he asks about every startup before anything else. "Is this a vitamin or a painkiller?" I turned this into a Claude prompt that runs his full product-market fit diagnostic on any business idea in 10 minutes. Here's how to use it:

Cowork changed the game for me. I run an Etsy shop with 48K+ sales and used it to rewrite all 136 product listings, build a customer service agent, and schedule content across platforms. Went from "I should learn to code someday" to shipping real automations in weeks. The skill system is where it gets wild.

After 2 months of using Claude Cowork daily, I can confidently say it’s completely transformed how I work. Here are 9 powerful prompts that automated most of my business—and they can do the same for you 👇 Bookmark this 🔖 before it goes viral.

I'd pay more. Running 3 businesses on Claude right now and the ROI is wild when you're automating real workflows — not just asking it to write tweets. Built a customer service agent, rewrote 136 product listings, automated content scheduling. If prices doubled I'd still come out way ahead.

The Claude Opus quality drama shouldn't surprise anyone. These companies have been deeply unprofitable, while hype fades and business fundamentals take over. The era of VC-subsidised cheap frontier AI is ending. Something has to give: either we pay more, or we get worse models.

The overthinking loop is so real. I went back and forth on pricing strategy for my Etsy shop for weeks. Fed Claude my sales data and competitor pricing — had a clear plan in 20 minutes. Biggest unlock was just dumping messy context into it and letting it find the pattern I was too close to see.

Whenever I'm stuck on a business decision for more than 3 days, I run these 10 Claude prompts and get more clarity in 15 minutes than 3 weeks of overthinking ever gave me. Copy and paste them into any chat:

There's a third group nobody talks about — people who couldn't do the work at all before. I'm a woodworker, zero coding background. AI tools let me build a customer service agent for my Etsy shop. I'm not replacing a top engineer. I'm doing things that were completely off the table for me 12 months ago.

An important nuance, not just restricted to design, is that the output of new AI tools will outperform the output of the *median* designer/PM/engineer today—so these tools will get popular—but a top designer/PM/engineer has a certain embedded context that AI doesn’t yet access.

@bdurangg Fellow Etsy seller here — 48,000+ sales, zero coding background. Started with AI listing optimization and now I'm building full agents for my businesses. The jump from "I can't code" to "I shipped something that makes money" is way shorter than it used to be.

Vibe coding ile iOS app yaptım, App Store'a attım, şimdi monetize ediyorum 👇 Kod yazmayı bilmiyorum ama uygulamalar çalışıyor. Bu benim için side hustle, keyfi bir iş. Ama keyfi olması para kazanmayacağı anlamına gelmez. Monetizasyon altyapısı kurmak ayrı bir dert normalde. App Store Connect'te ürün oluşturmak, fiyat tierlarını ayarlamak, RevenueCat'e bağlamak, paketleri tanımlamak. Bunları manuel yapmak saatler alıyor. 2 araç varmış, keşfettim. @egebese sağolsun 🛠️ RevenueCat MCP + Ascelerate App Store Connect CLI. Şöyle çalışıyor: Claude Code'a "şu uygulama için şu paketleri oluştur, yıllık şu fiyat, aylık şu fiyat" diyorsun. Ascelerate App Store Connect üzerinden ürünleri oluşturuyor. RevenueCat MCP paketleri tanımlıyor, offering'leri ayarlıyor, her şeyi bağlıyor. Bana kalan tek şey iki karar oldu 🤔 Paket fiyatlarını belirlemek. RevenueCat dashboard üzerinden paywall'u özelleştirmek. Gerisini iki MCP halletti. Vibe coding sadece uygulamayı yapmak değil artık. Build et, yayınla, monetize et. Tüm stack AI ile yönetilebilir hale geldi. Kullanmak isteyenler için ikisinin de linkini aşağıda mention'lara bırakıyorum.

I use Claude Code and Cowork daily to run 3 businesses -- Etsy shop, real estate, content. No engineering background. The wrapper startups gave me access to AI I wouldn't have figured out on my own. But now that Anthropic is shipping these tools natively, the value prop of "we made Claude easier to use" gets a lot thinner fast.

Anthropic just walked into the two fastest-growing product categories in AI and said "we'll take it from here." Prototypes. Lovable hit $400M ARR with 146 employees. Cursor is valued at $29.3B. Bolt, v0, Replit Agent. A $48B+ vibe-coding category built entirely on foundation model inference. Slides and one-pagers. Gamma hit $100M ARR with 52 employees at a $2.1B valuation, 70 million users, profitable. Tome, Beautiful.ai, Canva's AI slide generator all sitting in the same surface area. Claude Design ships prototypes, slides, and one-pagers in one product, powered natively by Opus 4.7. The economics of being an AI app wrapper worked because foundation labs were focused on model quality and left the UX layer open. That gap is the entire business model. Buy tokens at list price. Wrap them in a workflow. Charge 10x. When the supplier becomes the competitor, three levers remain. Brand distribution built before the lab shipped. Workflow depth the lab won't prioritize for years. Enterprise relationships the lab can't move fast enough to service. Everything else compresses. Gamma's 70 million users and cash-profitable operation is a real moat. Lovable's Fortune 500 deployments are a real moat. New entrants pitching a "better Gamma" or "better Lovable" as a seed round just lost the only arbitrage they had. The pattern rhymes with Microsoft and AWS. Own the horizontal intelligence layer. Pick off the verticals one by one. Code last year. Design today. Whatever ships in six months is already on the roadmap.

Learned this firsthand. Let AI write all my content for two weeks -- engagement dropped to zero. Switched to writing my own stories and just letting AI help with structure and editing. Night and day difference. The speed is addictive but it erodes the one thing that makes your content yours.

@AlexHormozi Built AI automations for my Etsy shop for 3 months straight. Broke everything constantly. Almost scrapped it twice. Now those systems run 3 businesses without me touching them daily. The moment I was closest to quitting was about 2 weeks before the first one actually worked.

Whatever you do, don’t be the guy who gives up at the exact moment when you should be fighting with everything you have. You’ll make it through either way, but there’s only one way you’ll look back and be proud of yourself.

@craigzLiszt This is literally how I run 3 businesses with zero employees. Claude Code builds my automations, Cowork runs my content pipeline, and I just point it at the next problem. The org chart is a joke until you realize some of us are actually living it.

refined startup structure: CTO: claude code COO: claude cowork CPO: claude design CEO: also claude

OpenAI is projecting $2.5 billion in ad revenue this year. So the tool I use to write customer emails and draft listings is about to start showing me ads while I do it. Weird time to be alive.

Context files are the unlock most people skip. I dump my entire business context into a project file — product catalog, brand voice, past decisions — and it goes from generic outputs to stuff that sounds like someone who works at my company wrote it. The prompt matters but the context is what makes it actually useful.

My most powerful vibe coding prompt ever. Plug this straight into the new Opus 4.7, and you'll literally be able to ship anything. Fully functional apps, landing pages, custom artifacts - you name it. Pro tip: plug this into a "vibe coding" expert project as a context file.

The pace is wild. Each upgrade noticeably changes what I can build as a non-developer. Stuff that took me a full weekend of prompt wrestling six months ago just works on the first try now. If Mythos is that much better I'm going to have to rethink what's even worth doing manually anymore.

As you use Claude Opus 4.7, keep in mind that however good it is, Mythos is still substantially better. The frontier is getting insanely powerful.

@bridgemindai The rate limit is a bigger bottleneck than model quality at this point. I split my work between Opus for the complex stuff and Sonnet for everything else just to stay productive. Shouldn't have to manage an AI budget like it's a utility bill.

Claude Opus 4.7 has been out for two hours and I'm already rate limited. 100% session usage. $200/month Max plan. New model, same problems. Anthropic shipped the best model on BridgeBench and you can barely use it. Some things never change.