Tag
This thread describes a secure and reliable development setup for agentic AI using a dedicated server accessed via SSH with tmux for session persistence, a terminal emulator with native tmux integration, and a VPN for secure access. The author advocates for this over local agent execution due to security and reliability concerns.