Ron Davis

Talk to an AI for 10 minutes and find your biggest gap

Update: it’s a coach now 💪 Thanks to the feedback here, the talk page got a big upgrade. Same link: Talk It Out What’s new: 🎙️ Daily coach: paste it once into a chat called “Coach.” Then every day just tap the mic and say “check in.” 3 minutes. It holds you to ONE next step, shrinks it if you missed, and parks shiny new ideas until Sunday. 📅 Sunday review: say “weekly review” and it goes back over your week and sets Monday’s step. 📚 Learn mode: say “learn” and it only lets you learn what fixes the problem in front of you. You explain it back in your own words and have to use it within 48 hours. One kick 10,000 times, not 10,000 kicks once. 📁 Coach as a folder: already run a second brain or AI operating system? 4 files you drop in, so the coach remembers across chats. (Credit @RyMac for this one.) Never done any of this? Start with the 10-minute interview at the top. It finds your biggest gap, and the coach keeps you on it. Try 3 check-ins this week and come back and tell us what happened. Missed days count too. Showing up to say “I didn’t do it” is the whole point. https://claude.ai/artifact/Rfnh6v91wkov3RFQHyrMSM

3 likes • 3d

@Mason Page the next-level thing is recognizing what success looks like for you and not trying to compare ourselves or be as good or as elevated as the people we treat as our mentors. Success to me means I get to spend time with my kid out on the river or the lake or at a theme park, so I'm at whatever it takes to get to that point. The rest trickles down from there. I saw your other post about the trades. It's an area that I work in. I think there's another member that I spoke to a couple weeks ago who's in Canada, also trying to work in the trades industry. Keep doing what you're doing. We'll support you. If you need help, reach out. I actually find myself teaching things less related to AI but more related to technology in general when I'm in my circle of friends. The funny thing is, I used to think they knew the same things I knew, but they don't. Or even if they know, they just don't have the time to spend on it to get it done at the level at which they want to get it done, so they never start. I used to think they had access to the same things I had access to, but they don't. Sometimes, when you start just exposing your knowledge and the way you do things to your circle, you find that there are people who need you, and you didn't know they needed you.

1 like • 3d

@Ry Mac same concept I would do.

Carla Bosteder

8d •

How to pitch a website?

I'm trying to figure out how to get more people to my website. It's not techy or an agency that builds for others. It provides encouragement and helpful information (products too) for women who are going through difficult seasons in life as well as some faith based content (gratitude, trusting God, etc.). I don't want to pitch to the folks who would use the material right now. I actually want to pitch to the people who know people who would benefit from the material. Does that make sense? Think: ministers, family counselors, support groups, hospice workers, etc. How do I do it in a way that doesn't seem salesy?

New comment 2d ago

0 likes • 3d

@Ry Mac I was just a witness to this this week with a lead, who happens to be a friend of mine. I gave him a great proposal to help him with a lot of his online social reputation. I gave him a very fair rate and time horizon to do the project, taking on most of the work myself. At least, I thought so, but he still had the perception (a valid perception, probably) that his team would have to have significant involvement. I really need to figure out how to pare down his needs to where he has a clear understanding that his team doesn't have to have that much involvement to get the benefit of this work. In a couple weeks, when I reach out to him for followup, I want to be able to show the limited scope that his team would actually need to be involved for me to be successful at doing the work. The point of this is that this situation is real. If they see your project as adding to their to-do list in some significant way, and they're not prepared to do that, in this case, he's got a low-staff situation. The prospect will say no, for now at least... Actually, one of the items of our project was to help him with his profile and job posting situation on LinkedIn, it was a small part, but a part nonetheless. Anyway He doesn't feel like he can take on more work until he can pick up the staff he needs to do some of the things he's engaged in. It's a revolving double-edged sword, which is dangerous.

Carlos Acuna

13d •

Playing around with Ry Mac's video pipeline

I took the ICM folder structure with a style I created, mixed it with @Ry Mac's video content pipeline and...Voilá! Excerpt taken from one of Ry's YouTube videos :).

Playing around with Ry Mac's video pipeline

2 likes • 3d

Okay, this is brilliant. Brilliant, just brilliant!

Carlos Acuna

9d •

I said "ok" 200 times. It cost 12 million tokens.

1 Chat, 1 Task: Reduce usage with clean context I guess most of you know that every message you send re-sends the whole conversation to your AI. Knowing it and seeing it are two different things, so I ran a test. One Claude Code session. I set a subagent to reply "ok" 200 times. One word in, one word back, every turn. Then Claude logged the tokens each turn actually sent. What came out: - 54,533 tokens before I typed anything. System prompt, my CLAUDE.md, rules, tool list. That block rides along on every single turn. - Each "ok" added about 54 tokens. After 200 turns the context was at 65,294. - 12.1 million tokens processed in total. For 200 "ok"s. The growth per turn is tiny here because the replies were one word. In real work, every file it reads, every tool output and every long answer stays in the history. Each turn adds thousands of tokens, not 54, and all of it gets sent again with your next message. Most of it comes from cache, but it still counts against your usage. So the rule I follow, and the one I teach: 1 task, 1 chat. When the task is done, close it. New task, new chat. If you need to carry something over, ask for a short summary and start the next chat with it. Full test page with the charts (you can hover every turn): https://claude.ai/artifact/4sFX5n6LANUnWDPRNonifM How long do your chats usually run before you start a new one?

I said "ok" 200 times. It cost 12 million tokens.

3 likes • 3d

This is an excellent model and a great way to portray it, and I'm glad I've been doing what I've been doing because it follows this expectation. I must have picked up on it early. For example, one of my ICMs that I use is in product management, part of software development, where we're gathering requirements, sorting through those requirements, and sorting through our technology stack to figure out how to meet those requirements. All that lands in a feature folder for that feature, with 12 stages of work to work through to get it to production-ready. Every morning starts a new session. Each feature gets its own session, and everything gets state stored into its feature folder and files in the ICM. Whether I start another session on that feature in an hour or next week, it's always up to date. It's actually quite remarkable how easy it is to stay organized now and how quickly I can answer questions I couldn't have answered in the past without significant research. Questions like ....did "so and so" actually say "blah blah blah" in a meeting about feature "abc". And claude with my ICM efficiently find the answer and gives it back to me.

Gabriel Azoulay

12d •

I tested my ICM folder against the document by itself. The document did the work.

Most tests of a folder compare it to nothing: a fresh chat with no folder and no document. The folder wins, and we call that proof. I ran the case that usually gets skipped. Same model, same nine questions, three arms: - No document, no folder: 1/9 and 0/9. - The document pasted in, no folder: 9/9 and 9/9. - The full folder in a Claude Project: 9/9 and 9/9. On three held-out questions graded by hand, the document alone and the folder both scored 3/3. The folder's only measurable differences were its section tags and one safety sentence. The accuracy came from the document. That changed how I build. The folder earns its place when there is routing to do: several stages, several files, a reader who has to find the right one. For one reference document and a question, paste the document. The repo keeps everything, including three discarded runs and why they were thrown out. The commit order is the proof: the questions and the scorer were committed before any arm ran, so nothing was tuned to the result. Try it yourself: pick one folder you're proud of. Before you claim it helps, run the arm with just its main document pasted into a clean chat. If that ties, your folder's job is routing, and you can judge it on that. Question for the comments: has anyone run the document-only control on their own setup? I'd like to know whether mine was the odd one out. Free and open: github.com/NFTYoginis/jdm-the-control

I tested my ICM folder against the document by itself. The document did the work.

2 likes • 3d

I'm definitely going to have to look more into this idea. I feel like a couple of my projects have potentially gone astray a little bit, and theoretically, this would bring it back into the fold or at least expose the gaps. Interesting!

1-10 of 59

Level 5

354 points to level up