Building learning tools that remember, learn, and work with you.

A working collection of evidence-based design patterns for learning experiences applied to page and screens and spaces.

Cross-client long-term memory — a worked example of how vector search, LLM extraction, and cloud functions combine to give AI tools persistent understanding.

A single control plane for a multi-vendor AI coding fleet — and a worked example of what breaks when five different AI tools share the same machine, credentials, and cloud services.

A local-first 2D colony monitor inspired by RimWorld — and a worked example of making parallel AI coding agents, context rot, and worktree collisions visible at a glance.

A personal research library searchable by AI — and a worked example of what makes semantic search actually good.

A daily AI/EdTech news digest — and a worked example of how to test whether an AI system matches your judgment.

The same fixed pipeline, now able to hand a project between Claude Code, Codex, and Antigravity — and a worked example of keeping the truth out of the model, the browser, and the conversation.

Agent pipeline for analytical reports — and a worked example of why constrained agents beat autonomous ones for long-running workflows.

A local-first document search tool for AI — and a worked example of when to keep everything on your own machine.