Infrastructure for human growth

A language learning framework combining two classic methods: Spaced Repetition Systems (SRS) for memory consolidation and Comprehensible Input (CI) for natural acquisition. I've already used a version of it inside Anki to self-learn German — the concept works, but the method of reproducing it hasn't been formalized yet.

The goal is to make language learning feel less like drill and more like the way children acquire: input-rich, curiosity-led, with memory handled by a system that knows when to surface what.

A multi-axis knowledge recording framework, inspired by kanban but built on a concept of information axes (dimensions of recording) that I developed myself. Kanban gives you one axis — status. Flexity lets you define and move across many.

The longer question it asks: how should knowledge be organized as state and movement, not just as text in files? This is thinking infrastructure, not a productivity tool.

A platform that generalizes and makes reproducible the training methods behind public drawing-skill curricula, with feedback made visible enough to keep going. It grows out of a simple observation: One can spend hours on a rhythm game and feel progress immediately, but struggle to put the same time into an instrument. Why?

The answer is feedback design — and the realization that almost any skill can be made game-like if its feedback loops are designed with care. Icarus is skill infrastructure: practice that wants to be practiced.

A program that works at the level of behavioral instinct to improve unconscious phone use. Not another screen-time dashboard — those just report; this tries to reshape the defaults that shape behavior in the first place.

It's the most direct expression of the agency value: in a world where attention is the resource being extracted, designing the conditions for your own attention is a form of resistance.

An exploration of combining the philosophies of Open Notebook and Obsidian into a single knowledge pipeline. Open Notebook handles the source side: automated collection, organization, indexing, and distillation of knowledge sources, with support for active recall and conversational learning. Obsidian handles the structure side: knowledge in reusable, LLM-native markdown, forming a bidirectional graph.

They correspond to two fundamental systems — the internalization of knowledge and the connection of knowledge. The goal is to automate the collection-to-learning pipeline so the human stays in the loop where they matter most: thinking, not filing.