Kyle Phillips

Working on some mesh subdivision + displacement w/ toxiclibsjs and #threejs then OBJ into #cinema4d #generative

Media gen is already living this correction, nobody stayed happy with prompt to image for long. ComfyUI, ControlNet etc all emerged to reclaim the feedback loop. There's still a ceiling. You can sculpt the output, but you can't think in the medium the way a painter can. Maybe that's where AI coding lands too, better than a mail slot, not quite hands in clay.

A strange thing is happening: I'm building an internet just for myself. I now have a whole network of self-hosted sites on my VPS that I visit throughout the day, they give me reports on my interests, show results on my projects, surface connections related to my thinking, display art that has been generated, present ideas to explore, and summarize news. Many of the things I used the regular internet for are now running on my own portals, designed for me alone.

Today, we're open sourcing the code behind "The Infinite Crate," our VST/AU plugin that lets you play with Lyria RealTime directly inside your favorite DAW! 🎧🎶 Since we released The Infinite Crate last year, it’s been used by some of our favorite artists - including a wonderful showcase with @daitomanabe in Tokyo - and being featured as a top new music tool at NAMM 2026. We’re now fully open sourcing the plugin for developers to fork, modify, and make their own under the permissive Apache 2.0 license. 💾 Get it here: github.com/magenta/the-in…

been having lots of fun playing with ASCII art and @GoogleDeepMind Gemini lately, really fun to see the abstract decisions that it makes

Why use petaflops for pixel when ASCII do trick? ¯\_(ツ)_/¯

@JohannesMutter this is so awesome, can't wait. godspeed!

I've been collecting interface designs for 15+ years. every gesture, tool, menu that made me stop and think “wait how does this work”, “why does this matter?”, but also “who are the giants upon whom we build?” Now i want to publish it: “Interface Lineage”.

@hikohiko Cc @alexanderchen :)

なんとなく作りたい作品のアイディアがあり、その技術的なテストとして、なんちゃって音声合成をやってみた。多摩美の授業で使っている自作の電子工作キットRabbit Pieを使った。

terminals can be such a delightful creative medium I created a terminal app for creating terminal apps - here's one prompt -> infinite generative ascii art applications, these examples all are using gemini 3 flash

I published my line-by-line review TUI for Claude Code (and standalone app). /lgtm-plan to try it Instead of reading Claude's plan and trying to prompt feedback after, Claude opens it in a split pane for you to leave comments on specific lines, and quit. Claude gets structured feedback instantly github.com/hapticdata/lgtm

the singularity will run on cron

@theo true... but tools are pretty essential, prompts are going to get you very far if it can't research any information via tools & mcps

You don’t need a bunch of MCPs and skills. You don’t need a customized TUI with tons of plugins. You don’t need custom sub agents and Ralph loops. You just need to prompt better.

@creativedash you have an excellent sense of how to prompt for motion design

Node Spawner: radial menu for creating nodes across many different categories. Click to open, scroll/swipe to cycle through.

@AndrewYNg @AnthropicAI @eschoppik love this partnership!

Important new course: Agent Skills with Anthropic, built with @AnthropicAI and taught by @eschoppik! Skills are constructed as folders of instructions that equip agents with on-demand knowledge and workflows. This short course teaches you how to create them following best practices. Because skills follow an open standard format, you can build them once and deploy across any skills-compatible agent, like Claude Code. What you'll learn: - Create custom skills for code generation and review, data analysis, and research - Build complex workflows using Anthropic's pre-built skills (Excel, PowerPoint, skill creation) and custom skills - Combine skills with MCP and subagents to create agentic systems with specialized knowledge - Deploy the same skills across Claude.ai, Claude Code, the Claude API, and the Claude Agent SDK Join and learn to equip agents with the specialized knowledge they need for reliable, repeatable workflows. deeplearning.ai/short-courses/…

✅ LGTM: I built a TUI and Claude Code Skills for reviewing AI-generated plans and code, line by line. lgtm gives you structured feedback types (blocker, concern, suggestion, question), vim-style navigation, syntax highlighting, and exports your review as markdown or passes it right back to your claude session.

Thrilled to launch Project Genie, an experimental prototype of the world's most advanced world model. Create entire playable worlds to explore in real-time just from a simple text prompt - kind of mindblowing really! Available to Ultra subs in the US for now - have fun exploring!

I've got Qwen3-coder:30b and gpt-oss:20b running locally in ollama but I am having trouble getting reliable tool calling with AI harnesses like OpenCode. Is there a different harness I should try? Would I have more luck with LM Studio? 🙏

GemBooth! test out prompts in Gemini 2.5 Flash Image (aka 🍌 nano-banana 🍌) with your webcam

Infinite Wiki ⁕ Every word is a hyperlink. Every description is generated in real-time (in ~1 second) ⁕ Runs on Gemini 2.5 Flash Lite. ASCII diagrams using 2.5 Flash

@alexanderchen very fun! incredible to think of how long this would have taken a couple of years ago

"Hello, world" 🖥️🌎 A short film inspired by the technology of my childhood, made with Veo 3. Sound on 🔊 🧵