Small, working demos of what current AI can actually do — mostly for language learners and students, the people I teach. Each one comes with a short note on how it works and what surprised me. Demos marked runs on your device never send what you type anywhere.
Ask a study question and watch Claude plan, call tools (searches over my real vocab decks, a calculator) and check its work, one step at a time.
Paste a paragraph from a college essay and get rubric-based feedback on voice, specificity, reflection, structure and clarity, with evidence highlighted in your own words. A look at where AI helps an admissions writer, and where it doesn't.
Paste any Japanese sentence and get furigana, a word-by-word gloss, and plain-English grammar notes — the explanation I wish I'd had on the Gotō Islands.
Read a Japanese sentence aloud. A speech-recognition model running on your own device writes down what it heard, and a kana-by-kana comparison shows where it heard something different.
Nine SAT-style question types, math and reading & writing, each able to generate endless new variants in a fraction of a millisecond. A lightweight open model, Qwen, explains any one on request.
Ask one question and three models answer side by side, from a 1-billion-parameter model small enough for a phone to a frontier model, with the time and cost of every answer.
Describe a scene and an image model paints it in ink wash, using the same kind of pipeline that painted Kitsune Run's world.
Language models don't read letters or words — they read tokens. Type English or Japanese and watch five real tokenizers carve it up, right in your browser.