Raymond Weitekamp

ypi. A recursive coding agent inpired by RLMs

VibeCAD. A library of Claude Code skills for teaching coding agents how to use various CAD tools

rlm-cli. CLI for Recursive Language Models

recursive-coding-agents. Recursive Coding Agents — AI Engineer World's Fair 2026 talk + website (SvelteKit, Cloudflare Workers)

MineTuning. Mine-tuning is a methodology for synchronizing human and AI attention.

dirpack. Budgeted directory indexes for AI coding agents with tree-sitter signatures

dspy. DSPy: The framework for programming—not prompting—language models

autoresearch. AI agents running research on single-GPU nanochat training automatically

longmemeval-rlm. Python

true-up. Deterministic, git-native 'truing up' for any repo: a content-hashed dependency graph that detects what a change makes stale, regenerates the mechanical, and worklists the advisory.

dynamic-agent-workflows. Canonical dynamic-agent-workflow patterns expressed across multiple agent frameworks — a cross-language teaching + falsification corpus. Runnable, smoke-tested, with teaching docs and per-language skills.

pkg-guard. Refuse npm/bun/pnpm/uv/yarn packages newer than N days. Defends against publish-and-detect supply-chain attacks.

studybench-lab-and-library. StudyBench lab/library environment manifests and results

fizzy-cli. TypeScript

ralph-eats-beads. Autonomous AI coding loop using bd for task management

blender-mcp. Python

voice-memo-hotwire-native. Swift

pi-lsp. LSP extension for Pi — code intelligence via Language Server Protocol

surrogate. Programmatic keystroke injection for agents to control any terminal application

pi-interactive-subagents. Interactive subagents for pi — spawn, orchestrate, and manage sub-agent sessions in cmux terminals

pi-reread-instructions. TypeScript