My setup now is pretty simple: claude.ai for working through problems, drive.google.com for files I actually want to keep, and floment.ai for the projects and tasks I'm actively executing. Models will change, prompts will change, probably half my current tools will change too. But having a clear place for knowledge vs actual work feels way more durable. Where does your useful AI output go after the chat is over?

Most people are still treating LLMs like goldfish with infinite context windows. But the real power comes when you give your AI systems persistent, structured, and reliable memory. I’ve been diving deep into three distinct approaches: - Open Brain (OB1) — the personal exocortex - Poor Man’s Memory (PMM) — the ultra-lightweight, git-native path @Millenial Cat - Cognee — the structured graph + vector layer for serious agents Each represents a completely different philosophy for how we should capture, store, and retrieve context. Full breakdown dropping soon: architecture comparisons, strengths & tradeoffs, how they actually fit together in a real stack, and when I’d choose one over the others. If you’re building any kind of long-term AI workflow, personal knowledge system, or agent setup — this one’s for you.What’s your current memory strategy? Drop it below

In yesterday's high tea, we discussed using a Claude skill MD FIle in another AI so that the agent can perform that skill itself, which it otherwise cannot do. In my case, I have a legal AI (not naming it to protect the identity of the guilty) that doesn't allow for much customization to get tasks done; everything is baked into its harness, or at least that is how it used to be. It recently allowed for matter folders to be created. So I had a crazy Idea, what if I took a skill i use in Claude for legal work and pull the MD Files and resources and create a specific skill/instruction folder to house these files and subfolders and use them in this Legal Ai. I gave it a shot and, in the chat, asked it to review the Skill/Instructions first and then take the other matter documents and execute the instruction file to create a specific analysis and document for an asset purchase I am working on for a client. For the first time, this AI analyzed the matter files and drafted a specific type of document that was almost the perfect format I wanted, with minimal editing. This one test has now opened the floodgates on what I can do to improve the outputs of this legal AI platform. I never even thought of trying this until yesterday's discussion over high tea. That one little gold nugget was a big win for me because it makes decent legal ai something more. Thanks @Jake Van Clief

Hey everyone, first update since my intro. In that post I said learning and building can quietly become a substitute for selling. 12 days later, I'm the proof. Here are my benchmarks as of today: 0 sales conversations with buyers, 0 paid clients, and 64 days until my deadline for a signed, paid client. That's on me. One thing I've taken from @Jake Van Clief is to ask yourself the hardest, most detailed questions you can about your own business and build toward the answers. When I do that, the hardest ones aren't about building. I built a pipeline dashboard on top of Karbon for an accounting firm with Claude Code. The hardest question I currently have is one a prospect could ask me on the first call: "Why you? What experience do you have that shows you can do what you claim?" Right behind it: "What is this costing me right now, in dollars?" I can't answer either one well yet, and until I can, getting people on a call and actually selling to them is going to stay hard. What I'm doing about it this week: sending my first 3 personalized Looms to small accounting firms that has the sole objective of getting a reply. No selling in the video. 3 this week to small accounting firms, then more once I see what lands. Fuck it, I'm not going to learn anything if I don't do anything about it. I'll post how it goes, even if nobody replies. What I'm asking for: if you've had real sales conversations for an AI consultancy or AI implementation business, I want to learn from you. I'm putting together a small peer-to-peer group that meets once a week. Everyone brings what they did that week, what they learned, and one thing they're stuck on, and we hold each other to it. Especially looking for people a step or two ahead of me, but same stage is fine too, as long as it's the same kind of business. Comment or DM me with what your business sells, who it sells to, and how many sales conversations you've had so far. Thanks. Khalil, of Norvain

I've spent over a decade building software for clients. Small project teams out at the client, with a big core product team behind us. A lot of you are building without ever having worked inside a software shop, so here's the habit from that decade I'd hand you first: https://youtu.be/CnRM0NccyXs Waiting for perfect requirements is how a project stalls out in analysis paralysis. So we didn't. We got the gist, built something rough, put it in front of the client, and then shut up and watched them use it. One client told us flat out they didn't want the imaging side of our system, the part that stores the scanned documents. Theirs was fine. So I built a rough demo: a form with its scanned document sitting right on it, no jumping into their imaging system to go find it. As soon as they saw it, the whole contract shifted to include it. No requirements meeting was ever going to get that out of them. People can't tell you what they want until they see it. That loop used to cost a team weeks per round. With AI it's an afternoon. Most people in the AI space use that speed to finish faster. I'd rather use it to be wrong faster, while being wrong is still cheap and everything is still running on fake data. The video also walks the part of the map most new builders skip after the loops: putting the pieces together, a full run from start to finish, real data with security on, rehearsing go-live. Going from loop one straight to production makes you the TIP guy: test in production. Don't be the TIP guy. Question for the room: who's your first user? If you're solo and building for yourself, who do you hand the rough version to, and how do you keep yourself quiet while they poke at it? P.S. Show your work bit: the slides were an HTML deck with click-through reveals, and the cut came straight off the Screen Studio click log. 27 clicks, 28 slide states, no timeline scrubbing. More soon on the architect/builder side of this, which is the same fresh-eyes idea run with two agents.