This is for people who are not technical. Someone with a normal job who saw something on YouTube and wants to be able to make something similar for practical use. Live with Danney Trieu · Four months ago I had never made a single AI video. I'm not a deeply technical person and never done any film in my life before, I'm just very passionate about this space. I figured out where the decisions live — which ones you hand to the machine, and which ones you take back. Then I found Jake & ICM just like many of you. Once I saw his method, everything got easier. More valuable than that, I found this awesome community that keep motivating me and supporting me. I like to give it back 🙏. Every Saturday, I'm hosting a Skool webinar here *UPDATE (09/14): Session records: https://drive.google.com/drive/folders/1TNk-42gZajrTWw-2tD3Fbn760UE-BeM_ Reelix explainer: https://danneytrieu.github.io/vibe-film-sessions/index.html Higgsfield affiliated link: https://higgsfield.ai/?fpr=nguyen61 My discord channel: https://discord.gg/EKab83ZGB Youtube: https://youtu.be/dryW_H54Rss ━━━━━━━━━━━━━━━━━━━━ Here's what I'm covering -> 🪜THE SIX LEVELS OF AI FILMMAKING 1 · PROMPT — you type it, the model decides everything else 2 · REFERENCE — you stop describing the look and start showing it 3 · MEMORY — your world lives in folders, so you stop re-explaining it 4 · BLOCKING — you decide where the camera goes, not the model 5 · DNA — your taste, written down, applied automatically 6 · AUTOPILOT — you point it at your work and it hands you the shot list Most people are stuck somewhere between 1 and 2 — and have no idea there's a 3. ━━━━━━━━━━━━━━━━━━━━ 💬 ONE ASK BEFORE SATURDAY Drop a comment below: which level are you on right now, and what are you trying to make?

just discovered (sorry if this is old news) you can browse the file tree in claude code, do we need VS code anymore?

Something I read in The Age of AI made me rethink how I use ICM. My natural tendency is to approach ICM through business process orchestration. I define the stages, describe the steps, and try to make the whole workflow explicit. But sometimes I take this too far and turn the stages into strict rules. The result becomes over-engineered, uses more tokens, and often works less naturally with AI. I recently had the opposite experience. We replaced a detailed process with a simple skill that described the essential pattern, and the results were much better. That made something click for me. Maybe stages should act as an interpretable scaffold, rather than a fixed algorithm. They can clarify the purpose, context, constraints, handoffs, and expected outcome, while still allowing the AI to work out the best path. AI may free us from having to define an entire process in advance. But if we specify less, evaluating the outcome becomes even more important. We need to check the quality of the result, capture what we learn, and use that learning to improve the scaffold over time. Does anyone else start with ICM and then gradually slip back into rigid, rule-based process design? How do you decide what needs to be specified and what should be left for the AI to infer?

I have been working on getting away from being a perfectionist for some time now. Many years. Sometimes I am successful. Sometimes not. Today was one of the not days. I was making many updates to my linkedin profile. I was doing them all at once for consistency purposes. I did not want updates going out to my network for every update that I was making. So I tuned off the “share updates” function. Unfortunately, linkedin can be sneaky. In some of the updates that I had made there was a separate share button that I did not pay attention to. My network received a bunch of updates. I especially did not want my latest employer to see. Not because they matter (at least I keep telling myself that), but because I wanted them to find out a different way. I felt sick when one of my friends on linkedin congratulated me regarding one of my updates. That meant to me that everyone got a boatload of updates. Looks like my whole network got only one update. At least that is what I know so far. I wanted the transition to go PERFECTLY. Well it doesn’t matter now. I just have to convince myself of that. 🥴 I guess what is done, is done.

I have been running a long-term memory store for my agents for a few months. It recently crossed a line: it is now large enough that rebuilding the index is something I schedule rather than something I do while the kettle boils. That changed how I have to think about it, and I want to know how other people handle the same corner. The problem is not size. The problem is that growth and usefulness are two different curves and only one of them is visible. The store gets bigger every day whether or not recall gets better, and nothing in it ever complains. Four things I do about it, for whatever they are worth. First, write down what a good recall looks like before you need it. I keep a small set of questions together with the answer I expect the memory to surface. When recall quietly gets worse, that set is the only thing that tells me. Second, date everything and let the older entry lose. Two entries that contradict each other are not a tie. The newer one wins unless the older one is marked as a decision nobody ever reversed. Third, keep what was decided apart from what merely happened. Most of what a memory accumulates is transcript. Very little of it is a decision. Mixing the two is what makes retrieval noisy, because the transcript always outnumbers the decisions. Fourth, make removal a normal operation rather than an emergency one. If nothing ever leaves, what you have is an archive and not a memory, and sooner or later you are searching an archive in real time. The part I have not worked out is this. When recall looks worse, I cannot say whether the memory declined or whether my questions drifted. The little set of questions ages alongside the store, and I have no good way to judge it. So here is what I am asking. If you run something that remembers across sessions, how do you know it is still earning its keep? Has anyone found a cheap way to show that, instead of a feeling that it seems to be working?