DeepStation

our team's first week of having both sol and fable the crazy thing is 30% of our cost was on fable

🎙 Join DeepStation Greece's launch session on "Solo Journey to a Final AI Product" featuring Myron Sampsakis-Bakopoulos, AI/NLP Engineer at Rollstack (updated to July 13th)! 💥 Everyone has the rough idea, but turning it into a real AI product people actually use is where most builders stall. Myron pulls back the curtain on the full solo journey, from back-of-the-napkin concept to shipped product. 🔸 The Real Challenge Moving AI from a promising prototype to something genuinely usable is the hardest part — a gap the industry still struggles to close. 🔸 What You'll See An honest, practical walkthrough of research, prototyping, customer discovery, testing, and deployment when you're building alone. 🔸 Why It Matters In the age of AI-assisted building, knowing what NOT to build and when to change direction is what separates ideas from real products. 💡 What you'll learn: 🔹 How to take an AI idea from concept to production. 🔹 How to find early users and validate fast. 🔹 How to balance speed with quality when building solo. 👤 Bio: Myron Sampsakis-Bakopoulos is a Senior AI Engineer at Rollstack, where he builds multi-agent pipelines that automate data analysis and presentation-deck creation. With a Master's in Machine Learning and Data Science, his work spans LLM and computer-vision fine-tuning, RAG systems, conversational AI, and complete production-ready AI systems. He has worked exclusively in startups, operating with high ownership and moving quickly from ideas to production. 📅 Date: July 13, 2026 ⏰ Time: 7:00 PM to 9:00 PM 📍 Location: Π55 Coworking Space Plateon 55, Athina 104 35 🎟️ Register here: luma.com/1awu3jxs Thank you so much for our DeepStation Greece's Founding Team who is making all of this possible! We're so excited for what we will do together to support and Greece's AI community:  🔹 Natalia Vavoula 🔹 Anni Panagiotopoulou 🔹 Apostolos Chardalias Building an AI product alone is daunting, but it's more possible now than ever. Come learn what it really takes. See you in Athens!

@ShokhzodjonT @OpenAI Great question! Bring your laptop (though you're also welcome to watch). We'll cover installation, but for the sake of time, would like as many folks to download beforehand as possible :)

@DeepStationAI @OpenAI do you need to bring your own laptop with codex already set up, or is that part of the evening?

🎙 @OpenAI Build Week Comes to Miami! Join us for OpenAI Build Week, an evening where builders, professionals, and AI enthusiasts come together to make something real with ChatGPT Codex. 🚀 No panels about the future. No vague AI hype. We’re going to build. You’ll see how modern teams are using AI coding agents to move from idea to working product faster than ever. We’ll build one app together live with ChatGPT Codex, then watch two real product demos showing Codex in actual builder workflows. 🎟️ Sign up here (limited seats): luma.com/itcgkxiy 💡 What You’ll Experience 🎁 Free OpenAI Codex tokens credited to attendees so you can build live with us during the evening. 🔹 Live Build Session — One App, Built Together We’ll create one working app using ChatGPT Codex and walk through the real workflow: idea → code → review → debug → UI improvement → working app. 🔹 Codex Demo 1 — Codex in the Wild A practical look at how builders are using AI coding agents beyond toy examples. 🔹 Codex Demo 2 — From Idea to Shipped Feature A hands-on product demo showing how Codex can help teams move faster across product, design, engineering, and iteration. 🛠 The stack we’re vibing with: OpenAI Codex · Next.js 16 · shadcn/ui · TailwindCSS 👋 Perfect for: 🔹 Anyone fascinated by turning ideas into working apps 🔹 Developers ready to build with AI coding agents 🔹 Designers curious about design-to-code workflows 🔹 Founders who need to ship MVPs faster 🔹 Product people learning how modern AI teams operate 🔹 Students who want to see real software creation up close 📅 Tuesday, July 21, 2026 ⏰ 6:00–9:30 PM ET 📍 The LAB Miami, Wynwood 🎟️ Sign up here (limited seats): luma.com/itcgkxiy DeepStation is an official OpenAI Academy Launch Partner with 6,000+ global members. We’re building more than AI skills, we’re creating economic development ecosystems where innovators are made. Proud to host this at The LAB Miami, one of Miami’s original homes for tech, entrepreneurship, and builders who actually build. @paw_lean @agnelnieves #OpenAI #OpenAICodex #MiamiTech #BuildWeek #AI #Startups #SoftwareDevelopment #DeepStation

🎙 DeepStation Academy launches Vibe Code Course: Boost Productivity with Claude Code & Codex — Skills, MCP, and Beyond!  This is a 4-week hands-on course taught by Grant Kurz, Founder & CEO of DeepStation with Katherine Burge, AI Enablement at Workday and previously at Google! Hosted at The DOCK! 💥 This course takes you from basic usage to a fully connected, automated, agentic workflow. By Week 4, you’ll combine everything you’ve learned into one real project you build and demo yourself. 🔸 Who This Is For This course is for people already building with Claude Code and Codex who want to go dramatically faster. 💡 What you’ll learn: 🔹 Week 1 — July 11: Foundations & Setup Get set up and learn to actually drive the assistant: operating modes, model and effort selection, permissions and flags, and both interface and terminal workflows. Close by building your own live automation. 🔹 Week 2 — July 18: Connect & Automate Build your own Skills, then connect via MCP to your database, Notion, GitHub, and project-management tools. Add CLI power and pre/post Hooks. 🔹 Week 3 — July 25: Agentic Workflows Start with spec-driven development, then delegate work across teams of specialized subagents, including design, content, and more. Call agents programmatically via API. 🔹 Week 4 — August 1: Capabilities & Capstone Generate real deliverables like presentation decks, then combine spec-driven development, subagents, MCP, and hooks into one applied build you demo live. 👤 Instructor Grant Kurz is Founder & CEO of DeepStation AI, a 6,000+ member global AI community that has hosted 100+ workshops with speakers like Sergey Brin, Founder of Google. Currently builds AI solutions for SMBs, universities, and enterprise. 👤 Instructional Design Katherine Burge does AI Enablement at Workday, turning complex systems into hands-on tutorials helping technical audiences get to “it works” faster. Previously a Senior Instructional Designer at Google. 📅 Dates: July 11, 18, 25, and August 1 ⏰ Time: 12:00 PM to 4:00 PM ET 📍 Location: The DOCK, 400 NW 26th St, Miami, FL 💬 Office Hours: Wednesdays, 6:00–7:00 PM ET, every week of the course 🎟️ Register now: luma.com/rspam3q5

🎙️ DeepStation is excited to share the post event summary and photos from our Vibe Marketing & AI Content Generation, a night on how AI is rewiring the way the best marketing teams work. Three speakers, real workflows, and the human taste behind it all. 💥 🤖 From AI Slop to Professional AI Content Systems: Grant Kurz (Founder & CEO, DeepStation) opened with how "slop" was Merriam-Webster's 2025 Word of the Year. His diagnosis: the problem isn't the model, it's that people automate before they know what they're trying to say. The fix isn't a better prompt; it's a communication blueprint. His thesis: "Don't automate content — automate a communication system." He walked through a seven-part blueprint (message, goal, brand voice, audience, channel, context, outcome), turned it into a reusable prompt template, then ran live demos turning ideas into channel-specific campaigns. 💡 Context is King: Sebastian Pinzon Duran (Founder & CEO, AIVO; ex-Publicis, WPromote, OMG) skipped the hype and showed what's actually working. After 15 years and $300M+ in media budgets across Visa, DIRECTV, Virgin Voyages, and HBO Max, his message landed hard: the marketers winning with AI aren't the ones with the biggest tool stack — they're the ones who've taught their AI who they are. He gave a live look at what ChatGPT and Perplexity already say about brands, proving AI visibility isn't optional. Context is the new edge. 🎨 Building Brand Worlds with AI: Mariana Lara (Founder, BOUND & AiLand; AI Creative Director) closed the night: anyone can generate content, but building a brand people instantly recognize is the real game. In a behind-the-scenes breakdown of one of her own AI-generated viral campaigns, she showed that AI's greatest value isn't replacing creativity — it's amplifying human taste, vision, and direction. Key Takeaways:  🔹 Context beats prompt tricks — teaching AI who you are is the new edge  🔹 Don't automate content — automate a communication system; the blueprint comes before the prompt  🔹 Anyone can generate content; building recognizable brand worlds is where human taste wins Thank you to our brilliant speakers, volunteers, and community for an unforgettable night!

We're opening the waitlist for our Monetization Gateway, which will allow you to charge for any web page, dataset, API, or MCP tool behind Cloudflare. The charges will settle in stablecoins over the x402 open protocol. cfl.re/4eUFdt6

Former Meta top-tier engineer: "I don't review the code anymore. I got to a point where I never catch anything the agents don't catch." He runs 20-30 agents at once and ships 20-40 PRs a day, work that used to take a full team a month. In 55 minutes he explains everything he knows and builds a fully working workflow from scratch. Watch it, then read the full guide on building loops below.

Fable 5 is back.

HTML mentioned

In the next version of Claude Code: subagents run in the background by default, so you can keep talking to Claude while your subagents work If you want your agent to run in the foreground, just tell Claude

Claude Fable 5 will be available again globally tomorrow. After a series of productive conversations with the US government, we're redeploying the model with a new set of classifiers to target and block more cybersecurity tasks. In the near term, some routine tasks like coding and debugging will fall back to Opus 4.8. We’ll continue to refine these classifiers over the coming weeks to reduce false positives and better distinguish genuine misuse from legitimate requests. We’ve also begun drafting a consensus framework—with Amazon, Microsoft, Google, and other Glasswing partners—for assessing the severity of AI jailbreaks and how AI developers should respond to them. We invite other industry partners and model providers to join us in this effort. Finally, we’re scaling up our collaboration with the US government on model testing and safeguards. This will include pre-release access to models and safeguards for evaluation, information sharing on jailbreaks and misuse, and dedicated resources for joint research. Thank you to our users for your patience, and to our partners across the government, industry, and the research community who worked alongside us to make Fable 5 available again. Read our full blog: anthropic.com/news/redeployi…

As engineering, product, design, DS, etc. melt into a new kind of role, I was reflecting on what roles might look like in the future. For example, when I look at the Claude Code team I see what I think is five archetypes: 1. Prototyper: comes up with brand new ideas; churns out many ideas, most of which don't ship 2. Builder: quickly turns a prototype/idea into production-grade product/infra 3. Sweeper: cleans up the UI, simplifies the code and system, unships, optimizes performance 4. Grower: takes a product that has been built and iterates on it to improve Product-Market Fit 5. Maintainer: owns a mature system to make it secure, reliable, fast, and efficient as it scales Many people span across 2 roles, and sometimes 3 roles. I also notice that these roles are not really tied to job function -- eg. across Anthropic, some designers match category 1, some 2, some 3; same for engineers, PM, DS. A healthy team needs a mix of these, depending on the product: - A product that is new and pre-PMF needs people that are strong at 1+2+3 - A product that is growing and has found PMF needs 2+3+4 and some 5 - A product that has strong PMF needs 3+4+5 and some 2 Maybe product roles of the future will look more like this, and less like the domain-specific roles of today?

How to keep AI spend flat while token usage grows exponentially: Not with friction and spend alerts. With better defaults, routing, and caching. Better Defaults (not Usage Caps) – Engineers can choose any model they want, but defaults matter. We’re experimenting with defaulting to open weight models like GLM 5.2 and Kimi 2.7 through our LLM gateway, while still encouraging engineers to choose the right model for the task. 91% of our employees were never hitting their usage caps, so instead of lowering caps and driving up alerts, we're moving to cheaper defaults. Note that code reviews use a diversity of models, so they can check each other's work. Better Routing – In our custom harnesses, we preprocess prompts and route to the best model for the job, considering cache hits and model pricing. For instance, you may want a frontier model for planning, but not for execution where they can be overkill. Ultimately, humans shouldn't be choosing models - AI can automate this task. Better Caching – Cache misses are the easiest way to drive your cost up. All of our requests are cache aware, so we’re reusing a warm cache wherever possible. For example, our cache hit rate went from 5% → 60% in LibreChat once properly implemented. Keep Context Lean – Start fresh sessions when switching tasks. Scope file context narrowly. Disconnect unused tools. Don't just compact. The goal isn't fewer tokens used, it's fewer tokens wasted. Better Visibility – Our engineers can use as many tokens as they want, from whatever model they want, but we’ve made usage visible – and the more you spend on AI, the more impact we expect. The goal isn't to suppress usage. It's to build the infrastructure that makes exponential growth sustainable. Putting this into practice has cut our AI spend nearly in half, while our token usage continues to grow.

🎙 Join DeepStation for "Building Brand Worlds with AI: How Visual Storytelling Becomes Marketing" featuring Mariana Lara, Founder of Bound & AILAND and AI Creative Director! 💥 Anyone can generate AI content, but can they build a brand people instantly recognize? Mariana breaks down the creative workflow behind her own viral AI video campaign and shows how visual storytelling becomes marketing that sticks. 🔸 The Real Challenge Generating content is easy; building cohesive brand worlds is the hard part. This session moves beyond one-off pieces toward visual identities people remember. 🔸 What You'll See A behind-the-scenes breakdown of one of Mariana's own AI-generated viral campaigns, from idea to recognizable brand world. 🔸 Why It Matters With trust, authenticity, and brand consistency now central to AI adoption, this is how creators stay faster, smarter, and consistently on-brand. 💡 What you'll learn: 🔹 Frameworks for consistent visual storytelling with AI. 🔹 How to build stronger brand recognition across platforms. 🔹 How human taste and AI creative direction work together. 👤 Bio: Mariana Lara is a Creative Technologist, entrepreneur, AI educator, and founder of AI Land and BOUND. Her work sits at the intersection of AI, art, visual culture, branding, and emerging creative technologies, helping artists, founders, and companies move beyond content generation to build recognizable brand worlds. As both a technologist and an artist, she believes AI's greatest value isn't replacing creativity — it's amplifying human taste, vision, and creative direction. 📅 Date: Tuesday, June 30, 2026 ⏰ Time: 6:00 PM to 9:30 PM EST 📍 Location: The LAB Miami 🎟️ Register here: luma.com/deepst-8bh1 This event is just days away — don't miss your chance to learn how AI turns ideas into brands people remember. Save your spot now!

🎙 Join DeepStation for "AI Content Generation: Authentic at Scale" featuring Grant Kurz, Founder and CEO of DeepStation! 💥 The best work often goes unheard, not because it lacks value, but because the people creating it are too busy to broadcast it. Grant shows how AI content generation closes that gap, letting any organization tell its story at the speed of its work. 🎟️ Register here: lnkd.in/e8pCm9VF 🔸 The Real Problem The world's attention flows to the loudest, not necessarily the best. AI, used well, finally lets the right ideas be heard. Used without care, it does the opposite, flooding feeds with slop and eroding the trust you've built, especially when there's no human in the loop to evaluate prompt templates and output. 🔸 Authentic vs. Slop Grant breaks down what separates true, on-voice AI content from the generic noise flooding the internet: amplifying a real story, never fabricating one. With 84% of consumers now demanding that AI content be labeled, authenticity isn't optional; it's the whole game. 💡 What you'll learn: 🔹 Generate authentic, on-voice content at scale: AI that sounds like you, not a chatbot 🔹 Define and lock your brand voice so every output reads on-brand across writers, channels, and campaigns 🔹 Build consistent, reusable prompt templates your whole team can run to ship faster without drift 🔹 Ground every piece in real context (your data, your sources, your receipts) so nothing gets fabricated 👤 Bio: Grant Kurz is Founder and CEO of DeepStation, a global AI community of 6,000 members bringing engineers, professionals, and enthusiasts together to stay ahead of the AI wave. He has hosted 100+ workshops and hackathons with speakers including Google Co-Founder Sergey Brin, ran the first technical workshop on the OpenAI Academy, and partnered with Google for multi-country hackathons. His work spans enterprise AI, factory automation for Coca-Cola and Stanley Black & Decker at Syscon, and building a startup analytics platform acquired by CoinLedger.io. 📅 Date: June 30, 2026 ⏰ Time: 6:00 PM to 9:30 PM EST 📍 Location: The LAB Miami 🎟️ Register here: lnkd.in/e8pCm9VF

🎙 Join DeepStation Greece's launch session on "Solo Journey to a Final AI Product" featuring Myron Sampsakis-Bakopoulos, AI/NLP Engineer at Rollstack! 💥 Everyone has the rough idea, but turning it into a real AI product people actually use is where most builders stall. Myron pulls back the curtain on the full solo journey, from back-of-the-napkin concept to shipped product. 🔸 The Real Challenge Moving AI from a promising prototype to something genuinely usable is the hardest part — a gap the industry still struggles to close. 🔸 What You'll See An honest, practical walkthrough of research, prototyping, customer discovery, testing, and deployment when you're building alone. 🔸 Why It Matters In the age of AI-assisted building, knowing what NOT to build and when to change direction is what separates ideas from real products. 💡 What you'll learn: 🔹 How to take an AI idea from concept to production. 🔹 How to find early users and validate fast. 🔹 How to balance speed with quality when building solo. 👤 Bio: Myron Sampsakis-Bakopoulos is a Senior AI Engineer at Rollstack, where he builds multi-agent pipelines that automate data analysis and presentation-deck creation. With a Master's in Machine Learning and Data Science, his work spans LLM and computer-vision fine-tuning, RAG systems, conversational AI, and complete production-ready AI systems. He has worked exclusively in startups, operating with high ownership and moving quickly from ideas to production. 📅 Date: July 9, 2026 ⏰ Time: 6:00 PM to 9:30 PM 📍 Location: Athens (Location to be announced) 🎟️ Register here: luma.com/1awu3jxs Thank you so much for our DeepStation Greece's Founding Team who is making all of this possible! We're so excited for what we will do together to support and Greece's AI community:  🔹 Natalia Vavoula 🔹 Anni Panagiotopoulou 🔹 Apostolos Chardalias Building an AI product alone is daunting, but it's more possible now than ever. Come learn what it really takes. See you in Athens!

🎙 Join DeepStation for "Context is King: Vibe Marketing" featuring Sebastian Pinzon Duran, Founder & CEO of AIVO and former Digital Media Director at Publicis and Wpromote for brands like Visa, DIRECTV, Virgin Voyages, and HBO Max! 💥 Everyone says AI is a must, but almost nobody shows you what it actually looks like in real work. Sebastian skips the hype and reveals the real workflows quietly rewiring how marketing gets done. 🔸 The Core Idea The marketers winning with AI aren't the ones with the biggest tool stack. They're the ones who've taught their AI who they are and how they think. 🔸 What You'll See A live look at what AI engines like ChatGPT and Perplexity already say about brands, whether marketers are watching or not. 🔸 Why It Matters As AI Overviews become commercial discovery surfaces, your brand's AI visibility is no longer optional. Context, built right, is the new edge. 💡 What you'll learn: 🔹 Why context beats prompt tricks every time 🔹 How to see what AI engines say about your brand 🔹 Real AI workflows you can use today 👤 Bio: Sebastian has spent 15 years mastering the algorithms that matter—first leading Visa's digital media across 14 markets in Latin America, then managing $300M+ media budgets at Publicis, Starcom, and Hearts & Science for brands like DIRECTV, Virgin Voyages, and HBO Max. He is now co-founder of AIVO, where he helps brands navigate AI search, building products, automations, and content systems every day with AI tools. Based in Miami with Colombian roots, and a musician on the side. 📅 Date: June 30, 2026 ⏰ Time: 6:00 PM to 9:30 PM EST 📍 Location: The LAB Miami 🎟️ Register here: luma.com/deepst-8bh1 The future of marketing isn't coming, it's already here and quietly rewiring how the best teams work. Don't miss your chance to see what's already working. Grab your spot now!

Introducing eve, an agent framework. 𝚊𝚐𝚎𝚗𝚝/ 𝚊𝚐𝚎𝚗𝚝.𝚝𝚜 𝚒𝚗𝚜𝚝𝚛𝚞𝚌𝚝𝚒𝚘𝚗𝚜.𝚖𝚍 𝚝𝚘𝚘𝚕𝚜/ 𝚜𝚔𝚒𝚕𝚕𝚜/ 𝚜𝚊𝚗𝚍𝚋𝚘𝚡/ 𝚜𝚌𝚑𝚎𝚍𝚞𝚕𝚎𝚜/ Like Next.js, for agents. vercel.com/blog/introduci…

The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Claude models is not affected. We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible. Read our full statement: anthropic.com/news/fable-myt…