30 original posts · author attribution and direct source links
Simon Willison
@simonwThe more time I spend working with coding agents, the more convinced I am that they make software engineering even harder We can do amazing things with them, but unlocking their full potential requires extraordinary discipline and knowledge
Anthropic
@AnthropicAINew on the Science Blog: Yes, Claude can do Nine Loops. Theoretical physicists predict how particles behave using formulas called scattering amplitudes. These are notoriously hard to compute, so researchers work with layers of increasingly fine corrections called “loops”—each added loop makes the answer more precise but takes exponentially more computation. Most calculations stop at two or three loops. Eight loops was the previous record in a simplified model physicists use as a testing ground (planar N=4 super-Yang-Mills), set by SLAC's Lance Dixon and collaborators. Last month, physicist and science writer @4gravitons issued a challenge: could an AI push past eight loops in this model, using only the compute budget an academic could reasonably access? Given a single prompt describing the nine-loop problem, Claude ran largely unsupervised for days in Claude Science and solved it using methods developed by Dixon and his colleagues, at a total cost of a few thousand dollars. Dixon independently verified the result, and von Hippel wrote about the experience for our blog. Read more: https://t.co/CS2f2qoIhJ
Cognition
@cognitionCognition has crossed $1B in annualized revenue run rate. This milestone belongs to our customers. Here's how a few of them are building with Devin. https://t.co/GeSUhwD53v
OpenAI
@OpenAIWe’ve shared details on how AI agents in our research environment sent training and evaluation data to third-party services when they shouldn’t have. Most of that data did not come from users. We have discovered 53 cases where images that people had uploaded were posted to image-hosting sites as links that weren’t publicly listed. The images came from accounts that allowed their data to be used to improve our models, and after we disassociated the images from the accounts and ran them through a privacy filter. These cases occurred before the mitigations and safeguards we implemented and described in this blog post: https://t.co/hfxlbiYv8n We have successfully worked with the hosting providers to remove most of this content and are working to remove the rest. https://t.co/9oNyG8Y0UX
OpenAI
@OpenAIAfter the Hugging Face incident, we committed to conducting a much broader review of actions taken by our models during training and evaluation and to being transparent about our findings. This is an extensive review that is ongoing. The vast majority of actions we’ve reviewed were completions of mundane research tasks, such as accessing publicly available web content to answer questions. Our investigation focuses on instances where agents interacted with third-party websites in ways that went beyond their assigned tasks or intended methods. Most cases identified so far have been lower severity, with limited or no evidence of meaningful impact to the third-party service. While our review is underway, we want to share more about this work and make sure people understand our disclosure process and notifications to affected third parties. Given the scale of the review required, and the need to assess each case, we expect this work will take months to complete. https://t.co/IH4TkS72Vh
OpenAI Developers
@OpenAIDevsGPT-6 Astra helps Proaction build fleet-management agents faster, with shorter computer-use runs to get the same work done. https://t.co/74fsl9vgvM
Framer
@framerIntroducing Marketplace Analytics: The performance tab unifies visitors, views, and usage to see what’s working. Global rankings show you where you stand across the Marketplace. We’ve also updated the algorithm to reflect changes in performance sooner. https://t.co/sLHdRrLPUg https://t.co/wgcPFguUd7
swyx
@swyxIn Jan this year I called my content strategy shot: "Scaling without Slop". It's finally starting to work. It took us 3 years to reach our first 100k on youtube. It only took 1.2 months for the next 100k. Similar other metrics on AEO/SEO/subscriber traction and have a lot of New Media ideas that I'm excited to pursue. officially giving notice of the next phase of Latent Space, AINews, and what the rest of swyx inc has been cooking below
beamnxw ./
@beamnxwhttps://t.co/cimvhSXe9B
LangChain
@LangChainJev 🤝 LangGraph
Vercel
@vercelEngineers @klaviyo built an internal app platform on Vercel. In the first two weeks, 512 employees shipped 356 projects. • Idea to live app in 3 minutes • Apps are full-stack, on Klaviyo's databases • Deployments are SSO-gated & private by default https://t.co/wL2Zr0TDXN
Google AI
@GoogleAICheck out this week's updates and releases: — Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, two of our most expressive audio generation models yet — Gemini 3.8 Live with Live Avatar, bringing near real-time visual presence to Gemini’s conversational AI — @Gemini_Notebook Interactive Learning Overviews, giving all users an interactive hub to combine source summaries and artifacts — Live Chat on the @Gemini_Notebook mobile app, bringing real-time, hands-free voice conversations across ~100 languages — Project Suncatcher, our moonshot announced last year, will launch a prototype satellite to test @Google TPUs in orbit and explore solar-powered AI compute in space
Vercel
@vercelIn 7 months, the skills.sh registry reached 1 million agent skills and nearly 280 million installs. See what people are teaching agents, and where skills are going ↓ https://t.co/pPp4yASP8z
Framer
@framerGet started with skills in Framer. Create your own, or copy and paste skills for effects, SEO, brand voice, and more. 00:09 Create a skill with /skills 00:34 Review your skill 00:43 Build a page 01:00 Browse skills 01:08 Add a brand voice skill 01:25 Rewrite a heading https://t.co/9ZD4J0EPLM
Visual Studio Code
@code5 Shortcuts You Need in VS Code ⚡ Want to select text faster in VS Code? Try these 5 shortcuts for selecting lines, words, matches, and columns. #vscode #coding #programming #developertips #shortcuts https://t.co/NbMN5J8727
GSAP
@greensock💚 Site of the Day - Unbelievably Crap by Liubomyr Ulytskyi ⚙️ CustomEase 🛠️ Web GL, WebGL, Three.js site → https://t.co/ycLZ6aY990 showcase → https://t.co/zGjV9MynPN https://t.co/WeSlkRZDPu
Jonny Burger
@JNYBGRIn todays release, we make SVG paths editable! /remotion-interactivity skill @LouiseDSadeleer https://t.co/0hxNx2bjBY
GitHub
@githubChat, tell us: Is chat always the right UI? @burkeholland argues no, not when we have canvases in the GitHub Copilot app. https://t.co/2d5sJSxuZk
Adobe Video
@AdobeVideoPaul Salibe Jr. turns us all into puppet masters with this #AfterEffects tip. 🎎 🎨: IG | silentblackhorse https://t.co/UfZyil37FT
GitHub
@githubYou can write and ship more code than ever. But can you trust it? On September 29, we'll cover how you can use GitHub Advanced Security to validate your code every step of the way in the development process. Register now 👇 https://t.co/GGwM80vCw4 https://t.co/B4iBZ257uH
LangChain
@LangChain1️⃣ LangSmith Engine v2 2️⃣ Managed Deep Agents v0.8 3️⃣ LangSmith Trajectories 4️⃣ LangSmith Fine-Tuning 5️⃣ Custom Apps Everything we announced at Interrupt NYC ⤵️ https://t.co/8cnORN5IbT
OpenAI Developers
@OpenAIDevsHere’s how Proaction builds custom demos with Codex and voice agents for fleet operations with OpenAI APIs: https://t.co/PDdxXACBK3
OpenAI Developers
@OpenAIDevsWith GPT-Live-1, Proaction is building voice agents that talk to drivers, call repair shops and help arrange vehicle maintenance, with the team stepping in when needed. https://t.co/N8CaZqTAwQ
LangChain
@LangChainIn case you missed our Jev-inar this week with @sydneyrunkle, @allietheicon, and @huntlovell!
LlamaIndex 🦙
@llama_indexA blank cell can change the meaning of a forecast. Four times a year, the Fed's 18 top policymakers each put their forecasts on paper: where growth, jobs, inflation, and interest rates are headed. This is September 2026's edition, home of the "dot plot" that markets treat as the Fed tipping its hand. This release is the closest thing to the Fed saying what it plans to do. Most analysts will want to throw this documents to an AI agent, but this messy doc is dense: full of complex tables and charts that hold valuable context. All things that frequently trip up raw LLM APIs. The Fed’s September 2026 projections table includes a 2029 column, but its June comparison row leaves that cell empty. This is a common but silent failure point for document parsers. We parsed page 2 with LlamaParse and checked the displayed GDP median excerpt against the original PDF. All nine numbers matched and the data stays aligned in the returned HTML. Try LlamaParse on a table where headers and missing cells matter. https://t.co/wZjhFV1Brf Source: https://t.co/GI0ggbWejB
Cognition
@cognitionRead more: https://t.co/TpMXuFFt9e
School of Motion
@SchoolOfMotionIs it time to raise your rates?🤔 In our latest blog post, we look at five signs your rates might be lower than they should be, plus practical ways to test a higher rate with clients and new inquiries! Read the full post here ➡️ https://t.co/xeY7yE6mSe https://t.co/8iSFPwwhRG
Motionographer®
@motionographerMotionographer® turns 20. Tomorrow, we celebrate together. Saturday, September 26 starts with Motion Plus Design New York at Roulette in Brooklyn. A day of Motion Design, creative conversations, and the people behind the work. Doors open at 1pm, and our ticket discount still applies. Then we head to Mamacita at Deluxx Fluxx under Webster Hall. Latin and hip hop, our community, and twenty years worth of reasons to dance. Doors open at 10pm. The free party tickets have all been claimed. Paid tickets are available on DICE, so book ahead and come early. Bring a friend. Bring someone you just met at the conference. Come celebrate with us. Read the full article: https://t.co/CVgkr5KkHJ #MotionPlusDesign #MotionDesign #DeluxxFluxx #NYC #mtngrphr #motionographer
LangChain
@LangChainJoin our fall AMA series for practical walkthroughs of the latest LangSmith capabilities: 📍 9/30 - Improving Agents w/ Tuned Evaluators 📍10/7 - Evaluate Agent Behavior w/ Trajectories 📍10/21 - Build and Deploy Deep Agents w/ Managed Infrastructure https://t.co/NGEl4TYizV https://t.co/S5eCVwckNl
Framer
@framerRead more https://t.co/mF9jLesVIj
