The Popover API puts your UI on a separate... top layer so you can stop fighting with focus management. Plus backdrop styling so short it fits here: [popover]::backdrop { background: rgb(52, 168, 83); backdrop-filter: blur(3px); } Build more with less code→show more

This was a fun experiment. I took a different... approach to make the coffee look more appetizing, and honestly, it made me want a cup while I was working on it. Create a hyper-realistic exploded vertical ingredient deconstruction infographic of a chocolate coffee drink. Centered vertical alignment, each component floating separately with clear spacing: cocoa powder particles (top), glossy dark chocolate sauce ribbons and droplets (separate layer), airy whipped cream mound (separate layer), smooth chocolate-coffee liquid layer (separate layer), roasted coffee beans (separate layer), sugar crystals (separate layer), and at the bottom an empty small transparent ribbed glass with a weighted base. Deep premium dark background: clean studio backdrop with a near-black graphite-to-charcoal gradient, subtle and smooth. Soft controlled studio lighting, crisp separation shadows under each layer, ultra-sharp macro realism, minimalist thin pointer lines and labels, high-end product infographic, 8K.show more

I saw that Cloudflare recently redesigned their UI and... noticed this really cool modal concept: some modals open to what appear as full pages, but are closable to return to where you left off This gives you focus and breathability by omitting the background distraction a typical modal with a faded/blurred backdrop would I recreated this style in Spamwise and ended up really liking it!show more

🖼️ You asked, and it’s here: Patron members on... Letterboxd can now choose their own backdrop for each film. To celebrate, we’ve teamed with Focus Features to unlock an exclusive set of customizable The Bikeriders backdrops for all members. Be sure to update your app versions releasing today. 💅 Read more about this new feature and how it works ➡️show more

A tour of the post-processing features in my WebGPU... engine. Chaining global passes like blur, bloom, and pixelate, on top of a disturb pipeline with its own options: distortion, black and white, lighten, darken, noise, RGB split, and more. It also supports layer and plane UV isolation. Planes can have optional shadows and rounded borders. Behind-the-scenes split view shows the raw pass vs final composite. Built from scratch with the WebGPU JavaScript API. Lightweight, minimal code, 0 dependencies, modular custom components 🧩. Next, vertex work and finish the MTSDF generator I’m building in Rust. ✨show more

Showers are such a small thing that can do... so much for you! The warm water and steam just relaxes you while you spend time with your thoughts 😌 Take a moment daily to breathe and focus on maintaining your peace, it does much more than you probably think 🌸show more

Hate being steep in your swing? Comment “steep” and... I will send you more drills to help with your game. Here is a drill to stop you from pulling on the handle in transition which will help lower the arms but also keep the club shallow and coming from the inside. So you canshow more

claude fable + seedance 2.0 finds 5-star airbnbs with... amateur photos. then it them into a cinematic ad and mails the host a postcard (fable is just the orchestrator. other tools handle the other tasks) the system you can sell to property managers: - it scrapes the market and keeps the 5.0s with barely any reviews - it skips anything a management company already runs - claude fable reads the galleries and keeps the ones the photos undersell - it pulls the real guest reviews and puts each quote on its matching shot - seedance 2.0 turns their own photos into a cinematic reel - white-label it and the property manager wins the door you can automate this whole process with your openclaw/hermes agent reply "REEL" + RT and i'll send you a free guide so you can build this too (must be following so i can DM you)show more

The shoot with Alexa was AMAZING!! She's a natural... 🤩. Here is a short reel to show you that i am not lying about it 😜(can you hear her moaning?) Alexa is always available to shoot with us. So send me your scripts and get your customs with this stunning woman! 🔥🔥🔥show more

Claude Platform on AWS is now generally available through... your AWS account. Claude Platform on AWS gives you access to Anthropic's native platform experience through your existing AWS account. Claude Platform on AWS complements Claude models on Amazon Bedrock, so you can access Claude through the approach that fits your needs. With Claude Platform on AWS, customers can: ◽ Access the full set of Claude API and console features with same day availability for all new releases and betas ◽ Use AWS Identity and Access Management (IAM) access control and CloudTrail audit logging ◽ Consolidate billing within their existing AWS accountshow more

✍️ I wrote an article on programming blur shaders... for beginners. You do not need to be adept at either math or programming to understand how it works :) Going from the most basic/straightforward type of blur, to what you see below, and more! Lots of pretty pictures, media, and interactive code to learn from. -- Box blur, with clipping and radial weighting. (So, no longer really a box blur...?) 🔍 #gamedev #indiedev #unity3dshow more

You asked for more time. We listened. The Kraken... CLI contest is extended to June 10. We built $DOG and $PEPE CLIs to show you how easy it can be. Now it's your turn to build with Kraken CLI. Short video. Post it on X, LinkedIn, or YouTube. Tag us so we don't miss it. Prizes: 🥇 $12,000 🥈 $8,000 🥉 $5,000 Full details:show more

Jan Desktop v0.7.8 is out 👋 You can now... run OpenClaw agents with local models, in just a few clicks. Plus Jan CLI for your workflows, and automatic context management so you never get cut off mid-conversation. Agent support is still experimental. We're working hard on it and would love your feedback 💛 Download Jan:show more

Jan Desktop v0.7.7 is live 💛 This update brings... native MLX support on macOS, a broader UX and UI refresh across the app, and better support for developer workflows. You can now upload files in Projects, use the local API server with both local and remote models, and work more smoothly with tools like Claude Code and other CLIs. Update your Jan or download the latest version atshow more

"Rankings outside of TOP 100?" YOU ASKED, I SHIP.... So many of you wanted to see your rankings outside of top 100, and I was thinking of the right way to do it without hurting Kaito AI 🌊 api server by requesting too much. and voila! now you can see rankings outside of top 100. I hope checking where you're at with the leaderboard motivates you to do more research and communicate with high quality mindshares! Seeing detailed progress on how much rankings you have gone up will definitely motivate you :D gKaito 🌊show more

1/3 🤖 Meet AgentOS: A Token-efficient, Microkernel AI agent... with on-device model routing across CLI, Web UI, and chat. A local router reads every message on your device and sends it to the cheapest model that can still do the job well. You stop overpaying for AI. ⚡ What makes it stand out: ■ Smart on-device model routing across 20+ providers ( @bankrbot LLM Gateway, OpenRouter, OpenAI, Anthropic, Ollama, and more). ■ Persistent local memory that survives restarts. ■ A layered security sandbox (Standard, Strict, Locked). ■ 37 built-in skills plus MCP, loaded only when a task needs them, and more ■ One unified gateway for CLI, Web UI, Slack, Telegram, Discord, and more Remember this: AgentOS has been integrated with the Bankr LLM Gateway since day one. That means any Bankr user with Bankr API key can start using AgentOS in minutes. Let the router cook.show more

The leading model runnable on a single cloud H100... GPU now fits on a single home GPU! 🔥 We've optimized Gemma 3 27B with QAT so you can run our best-in-class open model on your desktop RTX 3090 or similar. See how easy it is to try via ollama! 👇show more

What if your video pipeline worked like your codebase?... Introducing Replit Animation. Replit ⠕ #replitanimation From a single prompt, you can generate animated videos directly in Replit. It writes TypeScript with modern libraries to build scenes, apply styling, and export straight to mp4. Because it is built on code, you can import your real components, brand tokens, and design systems. That means consistent styling, reusable assets, and total control. Video as code means you can tweak, version, and scale your videos just like software. Create repeatable workflows for product updates, feature drops, and marketing campaigns without starting from scratch every time. Here is an example of a video I generated inside Replit. Export at your chosen resolution and frame rate or host with a unique URL. Activate video mode by typing video in your first prompt or selecting Animation under App mode. Create your first one:show more

MLP in PyTorch by hand ✍️ ~ 7 steps... walkthrough below Goal: fill in every blank in the PyTorch code to build a multi-layer perceptron. 1. Given Let us start with a code template on the left and the network it is supposed to build on the right. Every blank in the code can be worked out from the picture. 2. Linear layer We count: 3 features in, 4 features out. So the weight matrix is 4 by 3. There is an extra column for the biases, which means bias = T. 3. ReLU Let us apply the activation. ReLU crosses out the negatives, so -1 becomes 0. 4. Linear layer The input size is 4, because that is what the previous layer put out. The output size is 2. A 2 by 4 weight matrix, and this time no extra column, so bias = F. 5. ReLU We cross out the negatives again. 6. Linear layer Two features in, five out. A 5 by 2 weight matrix, with a bias column, so bias = T. 7. Sigmoid Let us finish. Sigmoid squashes the raw scores (3, 0, -2, 5, -5) into probabilities between 0 and 1. You have just implemented a three-layer deep neural network by hand. ✍️ == Story == Three years ago I gave this exercise to my students, to connect the code to the math. They found it odd. Every other AI course they were taking lived inside a Jupyter notebook, and here I was handing out paper. Three years later, my colleagues are the ones rushing to move their materials to paper. The exercise has not changed. Paper still asks the one thing a notebook lets you skip: do you actually understand what the code is doing? If you can tell me why the weight matrix is 4 by 3, and why bias is F on the second layer, you understand nn.Linear better than someone who has been copy-pasting it for a year. 💾 Save this post! #AIbyHand #PyTorch #DeepLearningshow more

I can't stop playing with Rapid Roboflow. The process... is so smooth that you can just: - Go to the rapid roboflow platform - Drop a few images or a short video - Type a prompt like “yellow cap” or “person wearing red hat” - Instantly get a working detection model No training, no GPUs, no cloud setup. Plug it straight into your app. It’s also baked into Roboflow Workflows, so you can instantly build automations like “count vials crossing this line” in minutes. Computer vision just went from weeks in seconds.show more

Gains with Eddie Pepperell 🔥 Working with EP on... laying the club down more efficiently in transition so that he can keep his sternum turning over his pelvis better in transition. He loves the feeling of being able to cover and get back on top of the ball with some rotation, so our goal here is to match his preferences & intent with a shaft pitch that offsets it. One of the best ball strikers in the game 👊show more