Sponsored
$5/week ↗

More projects built with Dots

Continue exploring projects, experiments, and ideas from the Dots community.

I Gave ChatGPT Dots a Real Job — Here's What Happened

I Gave ChatGPT Dots a Real Job — Here's What Happened

I gave ChatGPT Dots, OpenAI’s new always-on AI agent, a real job: handling the customer workflow for a painting business — quotes, owner emails, customer replies and bookings. Not a demo. Here’s what happened. OpenAI introduced Dots in September 2026. Dots are powered by GPT-6 Astra, have their own cloud computer and browser, and can keep working on tasks in the background. I wanted to go beyond the polished demos and see what happens when Dots has to handle actual business work from start to finish. In this video: → Setting up ChatGPT Dots for a real painting-business workflow → Testing the quote calculator — did it get the numbers right? → Drafting owner emails and customer replies with approval → Handling a booking end to end → Why some browser based tasks took 2+ minutes → My verdict: is ChatGPT Dots ready for real business work? ⏱️ CHAPTERS 0:00 Giving Dots a Real Job 0:55 Setting Up the Painting Business 3:23 Building the Painting Workflow 6:54 Test 1: Calling the Dots AI Agent 9:08 Test 2: Reviewing the Website Dots Built 10:23 Test 3: Testing Customer Enquiries 11:59 Test 4: Getting Live Project Updates 13:00 Test 5: Email + Calendar Booking Live 17:24 Final Verdict: Is Dots Ready for Real Work? 🤖 Want an AI agent like this built for your business? WebCrunch Systems builds AI automations for small businesses. For AI automation projects, collaborations, or business enquiries: gurmeetbuildsai@gmail.com 📌 Watch next: AI receptionist video 👋 I’m Gurmeet, a former Shopify senior software engineer with 18+ years of experience building software. On Gurmeet Builds AI, I test AI tools on real workflows — not just demos. 🔔 Subscribe for practical AI-agent builds, tests and real-world experiments. #ChatGPTDots #OpenAI #AIAgents
@GurmeetBuildsAI
Dot negotiated a denied refund back open and caught a missed promise from a 2h20 class

Dot negotiated a denied refund back open and caught a missed promise from a 2h20 class

Uma semana com o Dot e duas situações me chamaram a atenção. Na primeira, ele leu a transcrição de uma aula de 2h20 que dei no fim de semana. Encontrou uma promessa que eu tinha feito no meio dela, conferiu agenda e e-mails e veio perguntar pelo horário e link da próxima aula. Eu não tinha pedido essa revisão. Na segunda, conduziu um pedido de reembolso que a empresa havia negado. Pesquisou uma base para pedir reconsideração, preparou a réplica e negociou. Eu só fui aceitando o que ele sugeria, sem nem abrir meu e-mail. A empresa aceitou abrir uma exceção; o print mostra esse aceite e minha autorização para seguir. Naquele momento, ainda faltava processar o estorno. O que me interessa é ele voltar ao meu contexto, encontrar uma pendência e me trazer uma ação sem esperar que eu formule mais um pedido. Alguns agentes pessoais chamam essa revisão periódica de heartbeat. O Dot usa GPT-6 Astra. Minha impressão é que esse modelo pesa bastante na capacidade dele. Nos planos pessoais, ele começa no Pro de US$ 100/mês, com liberação gradual. Também há acesso no Business Premium e em beta no Enterprise. É promissor. Ainda está longe de ser indispensável para mim como um celular. Mas, entre os agentes pessoais que experimentei, está entre as melhores experiências de proatividade que tive. Compartilha o que o Dot fez por aí? Fontes sobre o produto e os planos: https://help.openai.com/en/articles/20001530-getting-started-with-your-dot https://help.openai.com/en/articles/9793128-about-chatgpt-pro-tiers
Dot "Boo" posts model-testing results to Slack and brings back replies over the weekend

Dot "Boo" posts model-testing results to Slack and brings back replies over the weekend

OpenAI's Dots launched last week and they've taken over @every. A few things that stand out: They protect your attention. Over the weekend, my Dot posted my model-testing results to Slack and brought back replies. I could keep working without opening Slack and disappearing down a rabbit hole. My weekend stayed pretty clear and calm. They're useful for the little things you miss. People at Every are using them to pull important information out of school emails, catch missed Slack messages, and flag account-access requests. Voice makes them useful away from your laptop. Our head of video was talking his Dot through video edits while running around getting our launch out. The ecosystem matters. Dots feel roughly at feature parity with Muse, Instinct, and Grok bots. But having an agent connected to the ecosystem you already use for work and life is a real advantage. If you use ChatGPT, I recommend trying one. Permissions are still frustrating. I have to approve the same basic thing multiple times, and sometimes even after I approve it, my Dot says it can't do it. That needs to get fixed. I'm less convinced we'll stay loyal to the characters. Mine is named Boo, and I'm attached to him. But I felt that way about my OpenClaw agent, too. When something better comes along, I switch. In a year, I expect to be using a persistent agent. Boo will probably be dead. RIP, Boo. I get into all of this at the start of this week's podcast, and then sit down with @bigwilliestyle to talk about where we go from here. In a lot of ways, Dots and GrokBots remind me of a wave we went through with OpenClaw eight months ago. We talk about personal vs. company agents, why we built our own company agent, and whether companies will end up with a whole shadow org chart of agents. Full episode below.
Projects

Dot designed and ordered Osaka business cards overnight — translated sites, placed the order, texted when ready

Despite how much I use AI agents in my day job, the launch of dots in ChatGPT for the first time has actually changed my day-to-day life. Just as an example, I'm right now in Japan for a week on a work trip. I had a long 12-hour flight from LA to Osaka. The Wi-Fi wasn't all that great, so getting work done would have ordinarily been difficult. But what I did instead was I just chatted with my Dot, told it the things that I wanted done, and got effectively a full workday out of that flight, just messaging back and forth as I simultaneously listened to podcasts and caught up on a bunch of talks I'd been meaning to review. I reviewed every side event, I prepped for every meeting. I even produced a brand new version of our sales deck, customized for every IRL conversation I was going to have. Then after getting to Japan, I realized I had forgotten to bring business cards. In Japanese business, exchanging business cards is very important. So I asked my Dot what to do. It figured out that there's a bunch of 24/7 print shops in Osaka that do business cards, it translated the websites for me, created a user account, designed the business cards referencing our Airfoil brand system, and placed the order. At no point did I open the website of the business card printing company or even open Figma. It then texted me when they were ready... and I just went to the print shop and my stack of business cards were there. The key thing here is when I'm doing design engineering or research work with AI agents, I am still very much in the driver's seat. I might hand off pieces of my work, but I'm guiding every small detail. Now that I have quite a bit of context saved and accessible for agents to use, and bluntly, agent harnesses and the underlying models themselves have just gotten so much better, being able to entirely hand off tasks and just trust that the agent will probably handle it (at least with the level of reliability of an intern) is a new feeling for me. At least so far, it does feel pretty liberating.