emi

agents-recap

Local, offline recap and RAG memory of your AI coding sessions. Reads Claude Code, Cursor and VS Code Copilot Chat session state read-only, tells you where you left off, and lets you ask about it later.

Python

m5-max-llm-bench

LLM inference on a Mac Studio M5 Max (40-core GPU, 48 GB): achieved memory bandwidth (≈550 GB/s, ~90% of 614 GB/s theoretical), decode vs bandwidth bound, mac-llm-bench results. Independent researcher.

Python

terminal-harbor

A GUI manager for many terminal sessions — built for juggling several shells and AI-agent sessions (e.g. Claude Code) at once. One large focus area with split panes, a sidebar of live thumbnails of every session, workspaces, and per-session "waiting for input" badges that tell you which terminal — a normal process or an AI agent — needs you.

TypeScript