I used Claude to map most of my files to obsidian but the token cost was crazy ($100+) because I insisted on using Opus- I wanted all concepts to be linked by meaning to other similar concepts. Ex- learning science to curriculum; science to biology etc. Since I was approaching my token limits, I let Antigravity 2.0 take over (Ultra plan) and it did everything without breaking a sweat. The entire wiki has more than 50,000 links and the output would be more than 20,000 pages to print 😓 The original plan was to use LM Studio and Anything LLM together to query the wiki, and that worked OK. Most of my stuff is research papers and documents. All that took a really long time to set up-big learning curve, and I learned a lot! But… Sometimes the easiest solutions are the least obvious. It hit me that I could just upload everything to NotebookLM and query that, which would’ve saved me 10 to 12 hours of work. Live and learn! What lessons have you learned or tips do you have for working with Obsidian?

Not which one you tweet about. The one diligently doing the work while you sleep. Mine’s a split household: Opus 4.8 makes the decisions, Sonnet 4.6 does the dishes. 🍽️ Pick your ride 👇

Who is building their own personal AI OS from the ground up? I would love to talk to you about your strategies and frameworks you are using. Im getting started with mine and plan to scale it to be a whole team running itself with an orchestrator.

I wanted to learn 80% of Hermes in 20% of the time. And I learn better by watching and listening. It's not perfect and I need to get Nous Research and many other phrases pronounced properly. Built and maintained by Hermes https://actionableops.com/hermes-howto

Keen to connect with anyone in the Seoul metro area. Thought partners, accountability, drinking buddies whatever! If you’re around hit me up!