Yi-Yi

I open-sourced Teachable Figures, a skill for AI agents to create beautiful, interactive 2D and 3D lessons. Try it here: github.com/ywEdAi/teachable-… It started with something I wanted for my own learning. I was studying AI model architectures and AI infrastructure. While reading about systems like SGLang, I kept wanting to see what was happening. Where does the data go? How do the components work together? What changes when I adjust one part of the system? My first version was a skill to help me understand SGLang through visual, interactive explanations. Then I discovered Anatomy and iso-figure. Their work on mechanical structures, cutaways, exploded views, and functional isometric objects gave me ideas for how much further this could go. I combined their rendering approaches with my teaching workflow. That became Teachable Figures. One question kept coming up as I built it: What kind of representation would help someone understand this particular idea? - For a Transformer, I want to follow the data through the computation and see how a causal mask changes attention weights. - For anatomy, I want to rotate a structure, look inside it, and understand where things sit relative to one another. - For a mechanical assembly, I want to separate the components and inspect their connections. - For historical clothing, I want to understand the overlapping panels, construction, and folds. Each asks for a different way of showing things. I also ran into this during iteration. A rotating model still needs a recognizable shape. A garment needs folds that communicate its construction. Choosing the right renderer and source material became part of the teaching workflow. Learning science helped me think through those choices. - Dual coding encouraged me to connect verbal and visual explanations. - Spatial and temporal contiguity shaped where explanations appear and how they stay synchronized with changes. - Cognitive load shaped what to reveal first and what to leave for later. The interaction follows a simple learning sequence: Predict → try → observe → explain → apply it to a new situation. Does adding another gear always make the output slower? If a cone’s cross-sectional radius doubles, what happens to its area? Does a rocket lift off whenever its engine produces thrust? I want learners to make a prediction, change something, and have a result they can reason about. The current demos explore Transformer attention, chest anatomy, rocket structure and flight, cone cross-sections, a Ming-dynasty robe, and gear ratios. They support Chinese and English and can run offline. My hope is to make this a tool I can reach for whenever I encounter something I want to understand, and let an AI agent help me build a small, beautiful lesson around it. The video shows a few examples. Try the demos: ywedai.github.io/teachable-f… Code and installation: github.com/ywEdAi/teachable-… Thanks to the creators of Anatomy and iso-figure for the work this builds on. If you try it, I’d love to hear what you’re learning and where the explanation falls short. Stars and contributions are welcome!

I’ve been building a little app called NotePico 📷 Take a photo of something around you, stick a word on it in a language you’re learning, and use that photo to review later. How it works: Take a photo → add a word you want to learn → listen to it → come back and review. That plant on your desk becomes a Japanese note. Your breakfast becomes a Korean one. A small way to bring language learning into an ordinary day. NotePico is ready for Android testing, and I’d love a few people to try it 🌱 Use it on something around you. Tell me what feels useful, confusing, or awkward. Reply or DM to join; send your Google Play email privately. notepico.space

Every AI tutor demo shows you the chat. Nobody shows you what the tutor is thinking. I built a visualization of what I think should be happening behind every reply: a working graph that grows as the conversation goes, and lights up one path of inference per turn. Content. The student's mind. Pedagogy. Representations. Tools. Five layers. One question: "why is summer hotter?" Watch the graph decide when to call a tool, and when to stay quiet and just ask. This isn't a product. It's a spec for one. If you're building tutors, this is the part that's missing.

An always-available partner solves an important availability problem. I think the next question is: what does the learner have a reason to speak about? I've been exploring two kinds of referential anchors: photos from a learner’s own life, and screenshots captured at any moment in a YouTube video. In the latter case, users don’t stop at image embedding or retrieval. They can use the screenshot together with its local transcript context to generate a small interactive learning artifact. For me, language learning and knowledge learning are inseparable. The goal is not simply to make learners chat more, but to make a meaningful moment something they can inspect, manipulate, connect, and revisit. Then speaking is grounded in an evolving understanding of something. Curious how you think about preserving that kind of learner-owned context across conversations.

his is interesting for learning systems too. A tutor shouldn’t only retrieve support for its explanation; it should surface the strongest counterexample or competing mechanism, then ask the learner to adjudicate. That turns retrieval from answer production into an activity of evidence evaluation.