Your curated collection of saved posts and media
And whaddaya know ... https://t.co/7P9hxp7rX6
Brendan Carr wrote the new textbook definition of jawboning when he said of Kimmel and ABC, “We can do this the easy way or the hard way." And he celebrates his censorship with a lame meme, which, in a sane world, would be an exhibit in a lawsuit.
And whaddaya know ... https://t.co/7P9hxp7rX6
https://t.co/Usy0E9tCmV
the pasadena wild parrots were having a party on our block tonight (sound on for the full experience) https://t.co/pGTVsImrku
@youwouldntpost Case in point jesus christ https://t.co/yjdaT0Iymj
@youwouldntpost Case in point jesus christ https://t.co/yjdaT0Iymj
Who’s joining us at the @MistralAI MCP hackathon this weekend? Still time to sign up if you haven't already! (Pro tip: We've hired straight from hackathons before… 😉) Huge thanks to our partners and sponsors @cerebral_valley @balderton @weights_biases @bria_ai_ @motierventures @qdrant_engine 🫶
How it works: - Open Playground → Evaluate. - Create or upload a dataset (CSV, TSV, JSON, etc). - Add Weave models: provider model + system prompt. Compare multiple models all in one run. - Once your prompt is ready, share it with your dev team so they can carry the prompts forward and build the end application using the Weave evaluation and evaluation-logger apis.
Evaluating LLMs in @weave_wb just got a whole lot easier! Now you can run your evals right in the UI, without writing a single line of code. Select datasets, pick models, include LLM-as-judge scorers, and hit Run. Simple as that. https://t.co/vfZLFyivTf
Get started right now. Like literally as soon as you click this link. First person to run an eval on W&B Weave Playground and screenshot it will get $250 in inference credits! https://t.co/nT9pcVdxlK
Evaluating LLMs in @weave_wb just got a whole lot easier! Now you can run your evals right in the UI, without writing a single line of code. Select datasets, pick models, include LLM-as-judge scorers, and hit Run. Simple as that. https://t.co/vfZLFyivTf
200+ builders showed up to our first WeaveHacks and the energy was electric! Now we’re running it back. 📍 wandb HQ in SF, Oct 11–12 - $15K+ prizes - 2 @UnitreeRobotics robot dogs - Theme: Self-improving agents Come build the next wave of agents that can upgrade themselves. https://t.co/puADZvdwDv
Huge thanks to the sponsors who make this possible: 🏆 Headline Sponsor: @googlecloud ⚡ @browserbasehq + @Stagehanddev 🖥️ @CopilotKit's AG-UI 🛠️ @mastra_ai 🚀 @daytonaio 🌐 @tavilyai They’ll be onsite with tools, feedback, and extra prizes. https://t.co/nrXDfZdgHj
🇬🇧 Fully Connected London Keynote Spotlight: @swtimmer, Senior Director, AI/ML and Decision Sciences @GSK, will take the main stage to show how AI code intelligence pushes agents from idea to production and share his journey. Join us in London, Nov 4-5. Details below! https://t.co/M6iQ10UY4v
Tickets are moving fast, get yours below! https://t.co/5tYiHuuuHx
“People think you can throw 100,000 humans at a problem… A lot of competitors are incredibly manual—basically body shops… literally dumping data into spreadsheets… there’s almost no technology going on.” - @echen, CEO & Founder of @HelloSurgeAI Full episode below! https://t.co/h8nVQCMiYJ
@echen @HelloSurgeAI https://t.co/HhDTCmRPMe
tldr: you can RL qwen3 8b to fool gpt-4o that it's not doing a hidden side task (when it is) this is somewhat surprising given the disparity in model capabilities between an 8b agent and gpt-4o as a relatively strong monitor https://t.co/TRefJQa8Rz
Announcing "look at your damn data" for Reinforcement Learning! You now get the full power of @weave_wb traces in the @weights_biases workspace. For any wandb run that calls an LLM, just "pip install weave" and "import weave" in your code. We also have out-of-the-box integrations for verifiers from @willccbb, ART from @OpenPipeAI, and verl from Bytedance Seed Team. Use those libraries with wandb/weave installed and everything just works.
Have you heard of spec-driven development? 👀 Check out our new Spec Kit, an open source toolkit to help you write specs that your AI can execute. 🤖 The result? Less guesswork, higher-quality code. ✨ Get the toolkit and guide.👇 https://t.co/bCMpKJwI18
Did you know you can turn your prompts into programs in Visual Studio Code? 👀 Create executable .prompt.md files for GitHub Copilot that you can run with a single click. 🖱️ Here's how. ▶️ https://t.co/csuj4KoZUM
Do you have an open source project or start-up that you want to feature LIVE on the keynote stage at https://t.co/lkLU54E9oL? I want to hear from you! Details here on how you can enter and maybe win a trip to San Francisco: https://t.co/4jQoHKY6O3 #ZeroToMergedContest https://t.co/qFpGHqn3Wm
GitHub Mobile is ready for iOS 26! ✨ What's new in our latest update: 💎 Liquid Glass visuals ⚡️ Faster navigation to diffs ✅ Improved multitasking Get more done on the go. Update now! 👇 https://t.co/VSABpHPhFL https://t.co/W7A62dYsaW
Made by @Grok Imagine https://t.co/ityQbZa8Vc
"Neuralink was like a gift from God" Powerful words from P7. I had the privilege of interviewing Jake Schneider, Neuralink's 7th patient, about what life is like now having a Neuralink brain implant for 3 months. Jake was diagnosed with ALS a few years ago and says before Neuralink, he felt overcome by hopelessness due to his condition. He says now @neuralink not only enhances his quality of life, but working with the team gives him a sense of purpose. And his son says his dad has superpowers now with his brain chip <3 My full interview is coming soon, but I was so touched by this story I wanted to share a preview. @PairedWith_P7
Grok Code Fast's 1M context window = CoT reasoning across entire repos! Per xAI's Aug 28 notes, I'm betting major upgrades: variants with multimodal vision for UI debugging & images, plus parallel tool calls for gnarly agentic loops. Screenshot-to-code soon? 👀 https://t.co/pZYJGNyZYb
Many major upgrades & fixes to Grok Code coming out next week, including 1M context window
I’m proud of this one. I brought a solar telescope to Florida to capture a Falcon 9 rocket launch transiting the sun. While these have been captured before, never with the details of the sun’s chromosphere, which makes this one the first! See the video or get the print below 👇 https://t.co/hFIo6vvXTU
@NanoBanana 🧵1/4 Adding and removing elements Using the provided image of [subject], please [add/remove/modify] [element] to/from the scene. Ensure the change is [description of how the change should integrate]. https://t.co/nynXjtEEQt
@NanoBanana 🧵2/4 Inpainting (semantic masking) Using the provided image, change only the [specific element] to [new element/description]. Keep everything else in the image exactly the same, preserving the original style, lighting, and composition. https://t.co/EHRuiQYJlq
@NanoBanana 🧵3/4 Advanced composition: Combining multiple images Create a new image by combining the elements from the provided images. Take the [element from image 1] and place it with/on the [element from image 2]. The final image should be a [description of the final scene]. https://t.co/3FFD7sin5Y
@NanoBanana 🧵4/4 High-fidelity detail preservation Using the provided images, place [element from image 2] onto [element from image 1]. Ensure that the features of [element from image 1] remain completely unchanged. The added element should [description of how the element should integrate].
