This isn’t a video. It’s a live, web-native prototype of work I designed for Trello. Real interface components. Real motion. Real interaction. It’s built in React and TypeScript—not rendered as a sequence of frames. I rebuilt the interface as live DOM so I could show the product work in motion, with panels, cards, and calendar elements layered independently in CSS 3D space. A custom browser timeline drives camera movement, rotation, depth, and pacing in real time, while responsive layouts and a Safari-optimized rendering path keep the experience smooth across devices. You can replay the sequence, switch views, and interact with selected parts of the interface directly on the page. It’s a front-end prototype, not a connected Trello product—but the interface, motion, and selected interactions are genuinely running in the browser. See it here:  All built with @OpenAIDevs Codex hosted on @ChatGPT Sites.

It has been an awesome few months at Trello! I have been designing several new AI experiences to help people stay focused, plan their work, and get more done, while adding a little delight along the way. You can now use AI in Trello to: • Find focus time for your tasks with Planner • Merge related items into a single card • Generate checklists that give you a plan for completing a task • Create an entire board from a single prompt • Bring more personality to your boards with AI-generated images And don’t forget to check out the Trello MCP, which lets you connect Trello with your favourite AI tools. It has been a busy and rewarding few months, and I have loved working with the team to design and bring these features to life. I am currently working on something especially fun that I cannot wait to share.

Now ready to try: I turned the GitHub repo into a live @OpenAIDevs GPT-Realtime-2 concept with @Replit so you can try it with no setup. Just open it, click the voice button, and ask about any stock, comparison, chart horizon, or market sentiment. I only put a few bucks behind the demo, so try it while it's live.

Been doing some travelling around Europe over the past couple of weeks. Got tired of aimlessly scrolling on socials while on buses, trains, planes (lots of waiting around) etc… So I made Pocket Alexandria. It works like a Twitter/X feed but for human knowledge. We have the world’s knowledge in our pocket and this experience creates an easy snack-size way to access it when you have a couple of minutes… or want to spend a little while wandering and discovering something new. Explore here: Has me thinking: what other tools can we build to harness and surface the knowledge trapped in these models? ps: built on @Replit and my phone while in transit throughout Europe.

I built a more polished follow-up to an earlier demo: a voice-first stock market terminal using @OpenAI GPT-Realtime-2, @threejs, and Codex. No traditional navigation. No clicking through screens. Just ask for a stock, a time horizon, a comparison, or context, and the interface responds. The UI is built around a 3D signal orb that changes with the conversation, revealing charts, prices, market context, and live data. This feels like where apps are going: voice-first generative UI interfaces that adapt to intent new experiences, not just new layouts The repo is public if you want to try it or build on it. Link below.

Using @OpenAI gpt-realtime-2 to get a glimpse of future voice-first experiences. A market dashboard you don’t click through. You direct it. Say, “Focus on Apple,” and the whole interface changes. Ask, “How did it do over the last 30 days?” and the chart updates. Say, “Go back,” and the market view returns. No menus. No filters. No hunting around. Just intent. What makes this model especially interesting is the interaction loop: you can interrupt it, add more context, change direction, and it keeps reasoning in real time while updating the experience around you. The interface doesn’t ask you to navigate. It just takes you there.