Tweet Signals

"Guys i think it's finally ready! You pick any area on a map and get a faithful, 1:1 3D version of it, in 30 seconds or less. It's tested, it works, and I'm shipping it this week once I'm done tightening the optimization. Right now it generates flawlessly anywhere in France and a…"

"V2. Brag+hyperframes+impeccable+make-interfaces-feel-better skills folded and consolidated into a single marketing-video skill with references. No guidance from me apart from "do a general overview video, take new screenshots via cdp, find some interesting things to show". It's …"

"The @nyknicks championship inspired me to build an AI basketball coach to improve my 1v1 game Built with: - Robloflow RF-DETR to detect players/ball/rim - MediaPipe to capture body angles - OpenCV to analyze and annotate the clip https://t.co/FSObOuekS2"

"your webcam is made of fabric: https://t.co/65G7Wq74Vy https://t.co/Gxx3LP9GXB"

"Today, we're launching Aside, the AI browser you’ve waited for. ⋅ Crafted in every detail: vertical tabs and Liquid Glass ⋅ SOTA on agentic browsing benchmarks: outperforms Claude Fable ⋅ Full privacy: Everything runs in local and encrypted ⋅ You can use Claude or ChatGPT subscr…"

"I had over 100gb of node_modules on my machine and I'm ALWAYS out of space use `npkill` to find and delete older node_module folders https://t.co/4jYbQb2F0V"

"Batchwork 1.2 released. Generate images up to 50% cheaper. https://t.co/5XCh7sGzui"

"I built a @NousResearch Hermes workflow that my team at @Box uses to track AI trends. Every day at 7am, it researches what's happening across the AI ecosystem and generates a brief for our team. We use it to identify: • AI trends worth paying attention to • Open source projects …"

"Release day: Clapet est en ligne! https://t.co/YzGBpQGd8j J'ai construit le tool de System Design que j'aurais aimé avoir pour préparer mes entretiens. Clapet te guide pas à pas pour construire une archi au lieu de te laisser seul devant un canvas vide a la Excalidraw. https://t…"

"As @FIFAWorldCup continues, take a look behind the scenes at @SawmillSports Hub in Chicago, where the research team at @RushMedical is using #Theia3D to study sport-specific movement tasks. Interested in using markerless mocap to study soccer movement? Contact us today to bring …"

"The harness can make a difference in AI for CAD. Below: same LLM (Opus 4.8) with different harnesses. Looking forward to see more progress without waiting for a new LLM version :) (Good job @pzfreo) https://t.co/nHTyPUCgCQ"

"One click animation from marker less video using MotionGen. #b3d #blender3d #animation https://t.co/BUGeKpJtmQ"

"3DTilesRendererJS suddenly hitting almost 35k downloads a week? Over 7x growth for the year - where is this coming from? https://t.co/PGQWT2cbEv"

"『移動型ロボットと人(選手)によるバドミントンの長時間ラリー(1,452回)』 このロボットは自律的に動き、弾道を予測し、正確にショットを打ち返す🏸🤖 https://t.co/HkimPiQ1X8 #badminton #mobile #robot #PhysicalAI #WorldRecords #GuinnessWorldRecords #GWR #ギネス #世界記録 https://t.co/Xs0S0eAR0z"

"I'm blown away. This AI filmmaking workflow for precise camera control, multiple characters, and dialogue is insane: 1. Generate a start frame in Midjourney 2. Match the poses in Blender, animate the camera 3. Feed both to Seedance I didn't think this would work. Two consistent …"

"1-bit GLM-5.2 GGUF vs. Claude 4.8 Opus vs. GPT-5.5 We gave 3 models the same prompt and compared one-shot outputs. The 1-bit GLM-5.2 GGUF ran locally on a Mac Studio M3 Ultra with 256GB RAM at ~21.6 tok/s. Which output do you like best? GGUF: https://t.co/BMkxswdj5N https://t.co…"

"Three skills I use every day in Claude Code and Codex to solve my hardest problems: 1️⃣ /agent-watchdog When I have one agent like Codex working on a task and I don't fully trust it's going to do everything right, I'll open up another one like Claude Code and tell it to watchdog…"

"What if your edge camera's video stream could reach any browser — with zero plugins and ultra-low latency? 🌐📷 We built an RTSP-to-WebRTC bridge for reCamera! On-device YOLO11n detection runs while video + inference boxes stream live to any browser via WebRTC. Perfect for dashb…"

"OpenAI’s new upcoming „bidi“-voice mode sounds insane! https://t.co/9UMzlCgEm9"

"Want to build TSL shaders without writing a line of code? TSL Graph is a visual node editor for the Three.js Shading Language compose shaders from nodes, see changes live on your meshes, and export clean, reusable TSL code. Perfect for learning and production. Check it out: http…"

"Object detection on 3D Gaussian Splats. Accurate bounding boxes per detection, runs locally. @nigelhartmanxr really cooked here @SensAIHackademy is hosting a lecture with @XRarchitect from @theworldlabs on the 8th on this: https://t.co/wOWO17Pw5G https://t.co/X5l0aMk2Sy"

"Arbor: Control 3D Generation with Explicit Geometry Text-to-3D makes plausible assets, but spatial control is still hard. Arbor adds typed 3D constraint: hull, avoidance, and touch. Paper w/ @markb_boss @AjlEngelhardt @Simondonne @CG_Tuebingen Intern @StabilityAI More 👇 https:/…"

"if you're running codex CLI, go check this path right now: ~/.codex/logs_2.sqlite ~/.codex/logs_2.sqlite-wal ~/.codex/logs_2.sqlite-shm since february, codex has been writing every single websocket event and telemetry trace to a local SQLite database at TRACE level. estimated wr…"

"京东最新开源的实时视频视觉语言交互模型：JoyAI-VL-Interaction，让大模型从“一问一答”走向“边看边说” 也就是说它会像人一样“在场”，持续观察视频流，自主判断什么时候说话、什么时候沉默，并实时响应关键事件 在58个真人盲评的实时流式场景中，对豆包胜率77.6%、对Gemini胜率 87.9%，监控预警场景对两个基线均100%胜率 可以用来搭建需要持续观察、自主判断、即时响应的实景AI，比如说安防监控、老人儿童看护、直播讲解、电商导购、操作指导、AI眼镜等等 核心特点是主动判断非被动回答，它会持续观察视频流，来自主做判断，不是等用户发起…"

"i'm obsessed with AI DIY projects my favorite one right now is this guy who built an AI system that listens for birds outside his apartment, figures out what species they are, and paints them on his wall. here's how the whole thing works: 1. a cheap usb mic on his balcony listen…"

"[1/4] Current 3D human pose reconstruction models are very impressive, but which model should you pick for your application? Introducing the Caltech Tennis Dataset (CalTennis), a large-scale video benchmark for evaluating monocular-to-3D pose estimation in the wild. CalTennis co…"

"Sakana Fugu surprisingly performed near GLM 5.2 level but 17× more expensive! We gave the same prompt to 4 models: build a complete live Trader Desk with both frontend and backend components, real-time market data fetched from external APIs for 8 symbols, and a custom dark-theme…"

"I've dictated almost everything for 6 months with Wispr Flow. 44,414 words, 161 wpm, top 0.1% of users. Last week I tried FluidVoice. Open source, runs local on my Mac, corrects as I speak with no API key, and handles slang better than I expected. Cancelled my paid plan. If you'…"

"New model release: LocalVQE Pi V1; A tiny 49k model that does acoustic echo cancellation, noise suppression and dereverberation at 21x realtime-time on a single Raspberry Pi 5 core. https://t.co/Fr1RtCiixd"

"This ComfyUI workflow lets you draw lines and annotations to guide motion, define where actions should happen, and influence movement throughout a scene. An integrated LLM translates those annotations into prompt instructions, making it easier to direct motion visually instead o…"

"Files SDK 2.0 just dropped. This is the full-stack release: one storage API from server to browser. → Gateway: 𝚏𝚒𝚕𝚎𝚜-𝚜𝚍𝚔/𝚊𝚙𝚒 → Client: 𝚏𝚒𝚕𝚎𝚜-𝚜𝚍𝚔/𝚛𝚎𝚊𝚌𝚝, 𝚏𝚒𝚕𝚎𝚜-𝚜𝚍𝚔/𝚟𝚞𝚎, 𝚏𝚒𝚕𝚎𝚜-𝚜𝚍𝚔/𝚜𝚟𝚎𝚕𝚝𝚎 → Server: 𝚏𝚒𝚕𝚎𝚜-𝚜𝚍𝚔/𝚗𝚎𝚡𝚝, 𝚏𝚒𝚕𝚎…"

"Still experimenting with hand tracking 🖐️ using a webcam and the dear Carlo. The 3D model was generated from a single image in @omma_ai Studio, and I remixed an Octopus demo to take advantage of the physics simulation. Building with AI these days is so much fun. Music also gene…"

"Can't believe 3D creators are sleeping on this. 🤫 If you’re a small indie artist or hobbyist you can probably use InstaLOD for free. We have our own remeshing features in KIRI Engine - but when we find gold, we love to share. Here’s a 3D scan we remeshed and re-baked textures f…"

"Whisper audio transcription on web (20x RTF): https://t.co/jXVv6sPnkS This demo was a breeze to make! The AI is good enough that I can just give it the jax-js repo + a .safetensors file, and it implements the model, builds a UI & makes inference fast Demo audio copied from @xeno…"

"Introducing Sakana Fugu: A full multi-agent orchestration system accessible via a single model API. Our ‘Fugu Ultra’ model matches the performance of Fable and Mythos, delivering frontier capability without the risk of export controls. Try it: https://t.co/hhO6qTawgb 🐡"

"This is one of the coolest open-source AI agent projects I've seen in a while: 'Understand Anything' It's a plugin for Claude Code, Codex, OpenCode etc. that analyzes your codebase and turns it into a knowledge base that you can interact with. It explains the codebase to you, ra…"

"Introducing agentcn 🤖 by @shadcnlabs > Built on Eve by @vercel and @flueai > Zero config, one command setup. >@shadcn /ui compatible (simply copy-paste) > 10+ production-ready agent recipes > Fully customizable 100% free and open-source. https://t.co/RSQO98he6R"

"💫Editing in 3D should feel as simple as moving a box.📦 Meet Thinking in Boxes: 3D Editing in Real Images Made Easy - fit 3D Boxes to any object in any image, drag/rotate/scale them or change the viewpoint, and get a photorealistic edit! 🧵👇 https://t.co/GeQcGF16jj"

"Web scraping will never be the same. (100% open-source visual search at scale) PixelRAG is a retrieval system that skips HTML parsing completely. Instead of scraping a page into text and embedding chunks, it screenshots the page and retrieves the image. A vision-language model r…"

"Interactive 4D Time-Lapse for #VDC in #Construction / #AEC Gauzilla Pro allows for creating a 4D time-lapse from a series of videos. You can convert drone footage into an interactive historical record in 4D running 100% on the web. ✅ 4D time-lapse drastically reduces the need fo…"

"Minimaps Saturday. Have now started experimenting with an option to pull Google 3D Tiles for a given supported area. Hilterfingen, Bern, Switzerland 46.74038°N, 7.65925°E Monument Valley, Arizona, United States 36.98512°N, 110.09987°W Meteora, Thessaly, Greece 39.71034°N, 21.629…"

"Bruh... this AI skill actually makes AI-generated UIs look better 🤯 Taste Skill is an open-source skill that helps AI coding agents generate cleaner, more polished frontends, and it's compatible with popular agents like Claude Code, Codex, Gemini, Cursor, Lovable, and more :) F…"

"COMPANY BEHIND TIKTOK JUST OPEN SOURCED AN AI AGENT THAT DOES YOUR WHOLE JOB FOR YOU China doesn't miss 😳 everyone's been crowning hermes the #1 agent then bytedance dropped deerflow 72,000+ github stars. 9,700+ forks. FREE. MIT it doesn't just run tools like hermes. it does th…"

"卧槽！秒啊！我怎么就没想到这个用法🤯 用 内嵌浏览器 做 Codex Image 2 的生图无限画布！ https://t.co/cMYwabZyWS"

"複数視点映像の特徴量を3D空間へ復元する「OneCanvas」。 1枚の360°パノラマへ再投影して既存の2D VLMに入力する手法。 複雑な3D専用モデルを用いず、SQA3DやVSI-Benchなどの空間理解ベンチマークで最高性能（SOTA）を報告。（詳細はリプ欄。） https://t.co/mHeRtj0YnR"

"SOMEONE VIBE CODED A POMODORO TIMER APP THAT USES YOUR AIRPODS TO TRACK YOUR POSTURE IN REAL TIME airpods already have motion sensors running while you wear them. no new permission, no extra battery, the data is just sitting there unused so he built a posture tracker on top of i…"

"This is GVHMR, a method for recovering world grounded human motion from monocular video. https://t.co/r5p8WLLoOf"

"Great touch by BBC to create a 3d looking at different views including 1st person. You can get a feel for a few things here with some scanning if the camera picks up the head moving https://t.co/si08QZaMmz"

"Open-weights models have officially caught up We tried GLM 5.2 in BrowserCode > Near Opus-level score > Cheapest model to date This task took $0.18 Try it now ↓ https://t.co/aR24ZDU1pj"

"Plannotator 0.21.0 is out Edit files. Co-context with an embedded cli agent. - Direct document editing - Launch an agent to build context with together - HTML files have first class support - Open any file you are viewing in external app - Bedrock/Vertex support - Stronger shari…"

"🚀MeshFlow: Efficient Artistic Mesh Generation via MeshVAE and Flow-based Diffusion Transformer | CVPR 2026 Highlight 📕 Project Page: https://t.co/ScyXHOl3fp 🖥️ Demo: https://t.co/5D18oW1e9C #3DGenAI #CVPR2026 #GenerativeAI #3DAI https://t.co/J7BlyRCc1L"

"Introducing CAD Drawing generation with Background agents. https://t.co/iqxaDHyJll"

"📢 OneCanvas: 3D Scene Understanding via Panoramic Reprojection We extract features from video frames and reproject them into one occlusion-free view of the whole scene that a 2D VLM reads just like a normal image. We can center this view on any viewpoint, including an agent's o…"

"Análisis táctico Inglaterra – Fecha 1 / England Tactical Analysis – Matchday 1 Revisión de los diferentes momentos del juego. ▪️Con balón ▪️Sin balón. https://t.co/yJxjTkT3bX"

"fine-tuning RF-DETR to detect hand heypoints still sucks, but I'll polish it over the weekend https://t.co/23OUJW6whw"

"today i'm launching Flashtype, a markdown editor for Claude & Codex use Claude/Codex to edit markdown, track every change, and see the diffs open source. available now. https://t.co/JtFc9m4Ae1"

"Latest update on the my #fifaworlcup app. 1) Shotmap 🎯 2) xG timeline 📈 Have a good and enjoy it👇 https://t.co/RzxS5e7IdX https://t.co/mcKQWVcQgG"

"We built a little tool that generates procedural soccer ball textures, one per country, for the World Cup ⚽ Made for the @threejsconf It exports all the maps too. Comment if you want the link ;) #threejs #webgl #vibecode https://t.co/QZuR9wT0Fz"

"Paid partnership with @NVIDIAStudio - You can now render entire CG movies with local AI in @ComfyUI on your own computer, for free. I built the whole pipeline with @ltx_io and used it to make a short film about a motel run by bears. 🐻 Free workflows + full tutorial 👇 https://t…"

"This is getting wild now 🫣 Just implemented a drone live stream into the iOS app from yesterday. Now I can get a bird’s-eye, top-down tactical view of a sports field additional to the iPad cams. Using @roboflow iOS sdk, vision models running at nearly 30fps on both streams. And…"

"We're open sourcing a 9B model that extracts structured data from documents at near-frontier performance. - 90.2% on our bench, vs Gemini 3.5 Flash at 91.3% - Leads extraction models like NuExtract3 (81.5%) - 9.5s p50 timings - Pass JSON schema https://t.co/zIdRbGjaY5"

"Update de FifaPhy: 1. Nuevas secuencias: paredes progresivas, cambios de carril a través de pivote y descargas a carriles exteriores 2. Funcionalidad scatter plot: seleccionas 2 metricas y ves las selecciones en el gráfico. Os comparto un video! Todo esto y mucho más en: https:/…"

"Loreline : un langage open-source et une app gratuite pour écrire de la fiction interactive, des dialogues de jeu vidéo et des récits à embranchements. https://t.co/OxDknj1SwU"

"Turned my handwriting in to a real font and used it in my designs (spotlight reveal font) 👇🏼 How it works: - Gemini finds each glyph - browser cleans the masks, traces them in to vectors. - opentype.js exports! https://t.co/gbLxKe4nXb"

"New Claude Code (v2.1.181) has a small but useful DX improvement You can now set any session setting inline with /config key=value, without opening the menu: /config useAutoModeDuringPlan=false Works in interactive, -p, and Remote Control Would like the same pattern for /model, …"

"holy crap, Lettera will be the de facto "markdown ide" within a year i've been using Bear 10 years, nothing comes close on markdown notes (yes, even obsidian) and now they're launching a standalone markdown editor https://t.co/9D0PkTFvIY"

"we ran opus 4.8 with text-to-cad skills against cadgenbench by @MikushRab and @huggingface and got some interesting results! the skills bumped the model performance by 0.04 points against the baseline (0.34 to 0.38) excited to run this again with fable 5 when it's back https://t…"

"Show Codex a workflow once. Reuse it as a skill. Record & Replay lets you show Codex a recurring task, like filing an expense report or submitting a time-off request. Codex turns that demo into an inspectable, editable skill. You control when recording starts and stops. http…"

"Launching today: No-slop AI slides at scale Lots of tools can generate a deck. Moda helps you orchestrate shippable assets - polished enough for customers, execs and investors. Upload a Google Slide or powerpoint. Add comments and lock key layers. Then let the agent cook. https:…"

"Build secure agentic workflows without iframe lag. Combining Agent to User Interface and Model Context Protocol Apps balances native rendering with custom iframe flexibility. 🛠️ https://t.co/bp8FtOOlwL 📦 A2UI over MCP for native layout 🎮 MCP Apps in A2UI for stateful modules …"

"LiteParse v2.1 is here, and its bringing the fastest markdown output possible. In this release, we are fulfilling our top request: markdown output. But in the spirit of "lite"-ness, we are doing this completely LLM-free and fast. Not only is it fast, it also beats all other mode…"

"ChatGPT can write you a thousand trading strategies from one sentence. 999 will quietly bleed your account dry. Every single one of them looks like a winner. One method pulls the real one out. Quant desks have run it for years. It's called loop engineering. It's not a smarter pr…"

"Video is an effective way to communicate, and we want to make it as easy as editing slides - that's why we created Google Vids (https://t.co/Z0lp7dvIRB). We are launching major enhancements to AI avatars, voiceovers, and video generation in Google Vids. Using these new updates, …"

"LFM2.5-ColBERT-350M is a surprisingly reliable smart tool selector. We gave it 151 tools, and it consistently surfaces the 5 most relevant ones based on the user prompt. This saves tokens and improves accuracy. Ideal for hmmmm agentic edge models? 👀 https://t.co/IPyizctesU"

"many of you asked whether this video was edited in Palmier Pro. here's the proof. try it free today. built for macOs. open sourced. https://t.co/Ala2PU4dk3 https://t.co/598f98pyPZ"

"https://t.co/r1PhRqrqaW 4.projem olan Football Analyst PRO Yayında! Maçları kendi oluşturduğunuz kod tablolarınız üzerinde hızlıca tagleme yapıp , sekanslar üzerinde çizimler yaparak , klipler oluşturup , temel istatistik haritalandırması yapabileceğiniz bir uygulama. https://t.…"

"a month from now you'll look at a comfyui output and have no idea what produced it the prompt is gone. the seed is gone. the workflow is gone. it's just a file in a folder i built a fiftyone plugin that embeds the full comfyui editor inside a dataset explorer run any workflow on…"

"3D scans usually begin as triangle meshes, which are great for capturing, viewing, and sharing real-world objects. For editing, animation, cleanup, or production workflows, quad-based topology can be much easier to work with. Our Quad-Mesh Retopology helps convert scan meshes in…"

"My dad’s acting as test subject again 🫣 70 yrs old. Former professional footballer. Still the GOAT ⚽️ Next step: LiDAR depth analysis distance goal ↔ ball. Runs smoothly on iPad. 👌 @roboflow @claudeai @GeminiApp https://t.co/hqwewtgSK6"

"Claude Code and Claude Design now sync both ways. Run /design-sync to pull your design system into your repo and build against your real components, or push what you've built back into Claude Design and keep editing on the canvas."