NEW Agent OS is INSANE! (Runs 3 Businesses FREE!)

Agent Operating System Q&A: Setup, Memory, Token Savings, Local vs Cloud, and Voice AgentsThis episode answers community questions about an Agent Operating System that keeps multiple AI agents and workflows in one place, including group chat orchestration, an idea-to-implementation pipeline, an agent Kanban, SEO and video agents, local engines, and a self-improving loop system, all supported by a shared memory layer. It contrasts the “old way” of juggling tabs, subscriptions, and lost context with the “new way” of a single, saved, synced workspace that reduces re-explaining. The host shares personal hardware and tool choices, explains that simpler setups can be best, and describes benchmarking models with a “Goldie Bench” to compare outputs. Additional topics include dictation tools, Hermes vs Claude trade-offs, ways to reduce token costs (free models/APIs/local), practical use cases, local GPU setups for high-quality generation, and options for Hermes voice agents.00:00 Agent OS Overview00:53 Old Way vs New01:36 Memory and Workflows02:03 Community Q&A Begins02:44 My Hardware Setup03:27 Paperclip to Pipeline04:11 Model Benchmarks04:54 Dictation Apps05:21 Hermes Troubleshooting07:04 Avoiding Overwhelm07:52 Cutting Token Costs08:50 Agent OS Use Cases10:03 Best Local AI Hardware11:55 Jose Customizes Agent OS13:22 Automating Substack13:53 Running Agents on VPS14:29 Hermes Voice Agents16:22 Laptop and Messaging Setup16:56 Wrap Up and Join