Robyn AndersonCEO, Jam Ai · AI Implementation & Business Development Leader

You don’t need to choose a single LLM anymore

9 Aug 2026 · 1 min

Watch on YouTube

1:03

In this presentation Robyn Anderson showcases her "AI second brain" built on Obsidian, which orchestrates multiple large language models (ChatGPT, Hermes, Claude, Codex) and specialized AI agents to run her business. She explains how each note in Obsidian is linked to rich metadata that directs AI agents, ensuring nothing is lost and no single model is required. A lead agent named Vicki coordinates the team, conducting regular stand‑ups so every agent knows its task and reports back. The core message is that the critical question isn’t which LLM to use, but how to design a system that integrates many models and agents to amplify productivity.

Chapters

This is my AI second brain, and I could use ChatGPT, Hermes, or Claude, or whatever I want to make sure that all of my AI agents are working exactly as they're supposed to. And I don't ever lose anything, and I don't have to stick to one specific AI tool. Every single little note that you see has so much information behind it and goes really deep, and it shows how they're all connected and how it all runs my business. If you haven't seen it before, it's called Obsidian. And when I'm working in Hermes, Claude, or Codex, everybody in my team as an AI agent knows what they're supposed to do and they report back. I have one agent named Vicki and she's in charge of the team, so she does team standups to make sure that everyone's on the same page all the time. So the question is not what can the large language model or AI tool do, The question is, what is the system you need to build to make these all so powerful?