My local model setup on an M4 Pro Mac mini

kept by devopsfolk sep 01

Kevin Lewis details his local LLM server on an M4 Pro Mac mini with 48 GB RAM, using Qwen and Gemma models via oMLX and Tailscale for private, cost‑predictable AI.

Nvidia’s Risky Business – Stratechery by Ben Thompson
Nvidia is partnering with major financial firms to create $500 billion in financing platforms for AI infrastructure, transforming compute into an investable asset class and expanding systemic risk in the AI buildout.
also on Apollo
hilos · team chat where people and AI agents share the room
Hilos introduces a team chat workspace where AI coding agents act as first‑class members, automatically creating branches, previews, and pull requests for human review.
also on Hermes, #ai-coding, #productivity, #developer-tools
GitHub - onecli/onecli: Open-source sandboxed agent harness for teams. Giving every employee a secured personal agent.
OneCLI is an open-source platform that gives each employee a sandboxed AI agent, managing credentials via a gateway and enforcing team policies.
also on Hermes, #ai-coding, #developer-tools
I vibe coded my dream macOS presentation app
Simon Willison describes creating a custom macOS presentation app using AI-assisted vibe coding in Swift, which sequences URLs as slides and includes a phone-based remote control via Tailscale.
also on Tailscale, #macos, #ai-coding, #developer-tools
Message your other Claude Code sessions - Claude Code Docs
Cross-session messaging in Claude Code lets sessions send text messages to each other using ListAgents and SendMessage tools, enabling coordination across worktrees, machines, and Remote Control, with delivery controlled by inbound settings and permission modes.
also on Tailscale, #ai-coding, #productivity, #developer-tools