Hey Jake, and everyone here, I want to say thank you, properly. I found this community back in March, off a YouTube short while I was away on holiday. When I got home I started learning everything I could about Claude Code, and built my first workspaces from the basics I picked up here. Then I just kept going: teaching myself, building, using it every day. It has genuinely changed how I work. I run AI and digital projects at a nonprofit (from strategy, governance and implementing AI projects), and the setup I built with Claude Code has reshaped that job day to day. I keep coming back here for the practical, hands-on resources while I go deeper on the AI side myself. The 7 mistakes post is a good example. The last few days, I ran my own setup against it. It turned out my always-on file was carrying far too much, I had a whole workspace I never actually used, and a few folders had quietly drifted out of date. I fixed all of it in one sitting. Leaner, and it just works better. What keeps striking me is how much the fundamentals matter. I keep finding resources here I'd never seen before, and now when I watch other videos, Obsidian setups and all the rest, I recognise the same structure, and I notice the people I actually trust all work in a similar way. I'm building my own web app at the moment, but honestly the biggest thing Claude Code has given me is a faster way to learn and sharpen my own skills. It's been incredible. Grateful to be here. Till

This is a wonderful and eye-opening community and I am glad to be here. I am working through my first detailed ICM agents, and I want to drop a few highlights, perhaps to be helpful, perhaps to get feedback. First, I chose an example that I knew well, so that I could judge the output fully. In my case that was seeding data from production into a new sandbox - I know old school SAAS salesforce. Second, this is 100% logical, detailed and has a right answer. Here is what I found. Claude really wants to show off all of its capabilities, and this makes it difficult to corral its capabilities into a logical path. In other words, I want Claude to stay within the md files, and follow exact instructions. This became notoriously difficult, as it seems to be pulled to discussing future steps like load order, or accessing the mcp, when the first stage is only about getting a list of objects/tables from a reference file. Yes, I got snooty with the AI. Yes, it was fun. And thanks to everyone who talked about Claude's memory because I found it difficult to have Claude stop saving to memory. Since I am dealing with larger data pulls, it also wanted to push that data into its .claude project location - so I had to be more specific about saving the data into output files than I thought. In other words, 'save the data output into the ftg-data-load.md 'file did not result in a list of records and fields. I have to ask it to specifically add all field information so that the file can be used alone. I have loved the path forward the past week or so. If the pace holds true, the absolutely positively critical points above will seem like basic thoughts in about two days.