Ken Burbary

Accepted into Built with Opus 4.7 🎉 500 builders selected globally by @cerebral_valley and @AnthropicAI for a one-week hackathon kicking off today. Grateful to be one. 🙏 What I'm making: 89% of AI prompts score D or F on the quality rubric I built six months ago. Most people have no idea. They blame the model, switch providers, pay more for 'smarter' AI, and get the same mediocre results because nobody ever told them the prompt was the problem. This week I'm publishing the proof. ✅ 500 prompts across 8 industries. Scored before Claude Opus 4.7 touches them. Scored again on what Opus 4.7 produces. Released openly on GitHub with a live demo anyone can use. First time anyone has mapped input quality to output quality at the model level. If the correlation is as strong as I think it is, it reframes how every company should be using AI. 🤔 #BuiltWithClaude #ClaudeCode #Claude #AI

The 'AI input quality' problem is real. It's getting louder. NEXT '26 was the week the stack caught up to the agent era and the quality layer fell one more beat behind. Full post: dev.to/onchainaiintel…

A2A is now running at 150 orgs. Long-running agents have permanent memory. Which means: LLMs writing prompts for LLMs, at scale, with no human in the loop, and nobody scoring the handoff. Observability tells you what failed. It doesn't tell you the input was the bug. 🐛

Google Cloud NEXT '26 shipped a full agentic stack this week. Studio. Registry. Identity. Gateway. Observability. Nobody shipped a quality layer for what flows through it. A control plane without a quality plane is the same shape as the early web with HTTP and no SSL. 🧵👇

In 5 years agents will be managing all those agents not humans but he's not wrong with the thesis.

@rye Your agents can already pay mine for prompt analysis and optimization using 'Prompt Quality Score' at pqs.onchainintel.net/for-agents 👋

Agents can now pay over HTTP. 75M+ x402 transactions last month. Almost none for physical products. Rye is closing that gap. Announcement soon.

@Nick_Prince12 x402 🚀🚀🚀

105,022 agentic buyers in the last 30 days 📈 agents are discovering services and paying in USDC no credit cards no API keys no human blockers what is your agent buying today?

@Must_be_Ash Just wait until you wake up surprised to a postcard from your agent one day. 🤯 Get well soon Ash!

gave claude a wallet to make a postcard and mail it to my friend Dan you just type what you want and it pays for everything with x402

ChatGPT Images 2.0 generated this. Hundreds of words. Zero typos. Legible at full zoom. 🤯 @Keshavatearth made it. 🙏 As someone who's spent 20+ years in marketing AND builds with AI daily, both sides of this land for me: 1️⃣ The content. It's the clearest single-page map of the modern marketing discipline I've seen. Foundations, GTM, Brand, Content, Paid, Demand, Retention, Analytics. The geography is correct. 2️⃣ The production. A year ago, this image was 40 hours of designer time and three rounds of revisions. Today it's a prompt and a coffee. The threshold we just crossed: infographic-quality visuals are now a writing skill, not a design skill. Great work Keshav and thanks @bentossell for sharing it! 🙌

@bentossell @Keshavatearth Edited for corrections. Thanks Ben. 👍

@kenburbary @Keshavatearth made this!

@Zain_Onchain @cerebral_valley @AnthropicAI So far it is terrific but intense pace. 3 more days.

@kenburbary @cerebral_valley @AnthropicAI Sounds good

Accepted into Built with Opus 4.7 🎉 500 builders selected globally by @cerebral_valley and @AnthropicAI for a one-week hackathon kicking off today. Grateful to be one. 🙏 What I'm making: 89% of AI prompts score D or F on the quality rubric I built six months ago. Most people have no idea. They blame the model, switch providers, pay more for 'smarter' AI, and get the same mediocre results because nobody ever told them the prompt was the problem. This week I'm publishing the proof. ✅ 500 prompts across 8 industries. Scored before Claude Opus 4.7 touches them. Scored again on what Opus 4.7 produces. Released openly on GitHub with a live demo anyone can use. First time anyone has mapped input quality to output quality at the model level. If the correlation is as strong as I think it is, it reframes how every company should be using AI. 🤔 #BuiltWithClaude #ClaudeCode #Claude #AI

@Jacobsklug 👋 here building as well. Welcome to Barcelona! 🤝

I'm in Barcelona for the month. Locked in. If you're around, have a private rooftop we can work at.

@0xyoussea Thanks for adding this part @0xyoussea 🙏 Just submitted the PR to add Prompt Quality Score (PQS) to the x402 universe. ✅

And you can contribute to the jsons adding your project here: github.com/youssefea/base…

My first time building with Opus 4.7 and I love it! I quickly built this Base AI Ecosystem map with names, links logos in the form of a galaxy that you can navigate through 🛰️ With some rigorous prompting, Opus 4.7 did everything for me including descriptions and logos!

@0xyoussea Visual network map, love it. Opus 4.7 is very powerful. Curious, was this a Claude Code output?

the 5 stages of ai grief since Claude Design launched, designers are grappling with the same existential recoil as when engineers first saw ai could code. the process maps to the stages of grief. 1. denial. "but design is more than just producing designs." engineers said the same thing. "coding is more than just writing code." both true. 2. anger. look how bad the output is. look at the people shipping slop. look at the execs who don't understand what we actually do. 3. bargaining. it's just a tool. i'll use it for the boring parts and focus on the strategic work. the craft is safe if i stay in charge of it. 4. depression. i can't believe i used to do all of this by hand. all those hours. all that time. 5. acceptance. i understand the nuance better than ever. i'm still the architect. and now i can actually build the thing. as a software engineer and designer of 25+ years, i've watched this cycle from both sides. the designers grieving now are where engineers were 18 months ago. when our core competency is threatened, we’re quick to defend what’s unique about it, romanticize it, and dig our heels in. what follow is a process of assimilation. i believe designers will eventually see Figma as an awfully archaic and cumbersome way to explore ideas. most designs already become interactive prototypes, so we'll just get there faster. much faster. in the end, taste and judgment is still what remains. creating successful work ultimately breaks down to a series of choices that add up to net value creation. those who win will continue to be involved in the most important choice-making, with a keen ability to discern between what choices are important for the human to make. think slow, move fast.

In crypto and defi (ie in honest markets), when a component fails, those closest to the component—whether wildly negligent or innocent victim—suffer the loss, and are burdened with that responsibility. Unequal, but proper. In tradfi and banking (ie in coercively manipulated markets), when a component fails, the entire society is forced under the burden of its resolution. Costs are socialized. Equal, but improper. The former, with time, becomes self-correcting, self-improving, and crucially, retains vitality. The latter, regardless of time, becomes stagnant and soulless, and here everyone can wallow in an equivalent grey. Any man of agency should prefer the former, taking care over that to which he is proximate. It is from this that the virtue of markets emerges.

@virtuals_io please check DMs. 🙏