Hermes Agent Explained
The transcript introduces Hermes, an AI agent that retains memory across sessions unlike standard LLMs. It highlights key features including cost-efficient reasoning via Open Router with Quen 3.6 Plus, Obsidian integration for persistent context, and cron job automation for token-free repetitive tasks. The entire stack is noted to run on an Android phone.
Summary
The transcript opens by contrasting Hermes with OpenClaw (likely OpenAI's ChatGPT or similar), criticizing the latter for lacking persistent memory — metaphorically described as a third date where your name is forgotten. Hermes is positioned as a superior alternative that remembers context across sessions.
The setup process is described as straightforward: copying an install command from the documentation and running the Hermes model in a terminal. A key cost optimization tip is pairing Hermes with Open Router and the Quen 3.6 Plus model, which the speaker claims reduces token costs by roughly 90%, turning a $100 spend into $10 while still delivering top-tier reasoning.
Hermes' standout feature is its memory layer, which audits its own past successes and carries forward relevant context rather than starting each session fresh. This enables it to function as a personalized command center. When integrated with Obsidian (a note-taking app), the user's notes effectively become the agent's brain, allowing it to sync plans, tasks, and context into a self-organizing dashboard.
For repetitive workflows, the speaker recommends defining logic once and scheduling it as a cron job — a locally executing script that runs automatically without making any LLM calls, thereby burning zero tokens. The transcript concludes by noting that this entire stack can run on an Android phone, emphasizing its accessibility and portability.
Key Insights
- The speaker argues that Hermes' memory layer audits its own past successes and anticipates future needs, rather than starting blank each session — a fundamental differentiator from standard LLMs.
- The speaker claims that pairing Hermes with Open Router and Quen 3.6 Plus reduces token costs by approximately 90%, converting a $100 token spend into roughly $10.
- The speaker describes Obsidian integration as turning the user's personal notes into the agent's brain, enabling it to sync plans, tasks, and context into a self-organizing dashboard.
- The speaker explains that cron jobs allow repetitive logic to run as local scripts on a schedule with zero LLM calls, meaning no tokens are consumed for automated recurring tasks.
- The speaker asserts that the entire Hermes stack — agent, memory layer, Obsidian integration, and cron automation — is capable of running on an Android phone.
Topics
Transcript
[0:00] Imagine a third date where they ask you your name again. That's how little OpenClaw knows you, but Hermes is different. To set it up, simply copy paste the install command from their docs and run Hermes model on your terminal. Now, your first power move is to pair Open Router with Quen 3.6 Plus, which instantly gives you top tier reasoning at a fraction of the cost, turning a $100 token burn into $10. But what's even cooler is its memory layer. Instead of starting blank every time, Hermes audits its own success, pulling [0:31] forward what worked in the past and anticipating what you'll need next. That's what allows it to become a personal command center, especially…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Greg Isenberg
What are Agentic Loops?
Professor Ross Mike explains agentic loops — AI systems that self-iterate without human input — and argues they are largely impractical for most developers due to high token costs and poor assumption-making. He distinguishes between 'human in the loop' (iterative, guided development) and fully automated loops, recommending loops only for constrained, binary-feedback processes like code review. He shares his own GP loop use case as a rare valid example.
Clearest Explanation of AI Agents as Customers
The speaker argues that AI agents are becoming the primary 'customers' of the internet, replacing human users as the dominant force driving web traffic and commerce. This shift requires entirely new infrastructure — from agent-native inboxes and wallets to machine-readable websites — creating massive startup opportunities. The speaker frames this as a bifurcation of the internet into a human layer and an agent layer.
How to use Obsidian with Claude in 61 seconds
This short tutorial outlines three levels of using Obsidian with Claude AI to accelerate professional learning and career resilience. It covers connecting Obsidian to Claude via MCP, using the Smart Connections plugin to identify skill gaps, and prompting Claude to generate a public learning plan based on your notes.
Been playing with this cool ai tool from genspark
The video demonstrates how GenSpark's AI tool can automate LinkedIn job post lead generation by identifying decision makers and drafting personalized cold emails. It then presents a business model called 'Outcome as a Service,' where users build competitor-tracking workflows and sell daily briefings to clients for recurring revenue.
How to make really good Claude skills (clearly explained in 42 seconds)
The transcript outlines a method for creating high-quality Claude skills by providing context, defining clear criteria, iterating until successful, and letting the AI write the skill itself. The process emphasizes testing and collaborative debugging to ensure reliability. The speaker argues AI-written skills outperform human-written ones because they reflect what actually worked.