User
Write something
Pinned
🏆 HOW COMPETITIONS WORK FROM NOW ON 🏆
Quick update on the competition schedule so everyone knows what to expect. 📅 NEW CADENCE: TWICE A MONTH We're dropping comps on the 15th and the 30th of every month. Two chances to compete, every month, on a set schedule you can plan around. ✍️ WHY THIS SCHEDULE Spacing them out this way means we can give tailored feedback on every single submission. Not just the winners. Everyone who enters gets notes on what worked, where it's weak, and what to do next. 🎁 WHAT WINNERS GET Along with the prize, every winner gets a 15-minute one-on-one with Jake. Use it to talk through your build, ask questions, or bring whatever else is on your mind. Two comps a month. Feedback on every entry. Direct time with Jake for the winners. Mark your calendar for the 15th and let's get to work!
Pinned
🎆 GOOD NEWS: THE SALE STAYS OPEN. HAPPY 4TH 🎆
We're holding the last sale through the holiday weekend so nobody misses it. 🎉 Premium: $27 → $14/mo 🎉 VIP: $97 → $67/mo This is the cheapest it will ever be. Once it closes, the price is gone for good. ⏰ New deadline: July 5th, 10:00 AM EST. This is the last extension. If you've been on the fence, sign up now. You lock this rate in and keep it every month going forward. 🖥️ ONE MORE REASON TO JOIN The week of July 5th we're dropping the software we've been building for this community. It goes out for beta testing first, and only Premium and VIP members get access. Sign up before the sale closes and you're in from day one.
Tracking moving parts 🧰
Question for the group: has anyone used AI to streamline managing project tasks like material ordering, POs, and tracking requested vs. received items? As a PM, I find it's becoming overwhelming. Currently, I track everything in Excel, which works, but it seems like there should be a more efficient approach. I even tried creating my own "Jarvis" to monitor it all, but it's still rough. I’m curious if anyone has found a solution—whether it's a custom tool, an existing product, or a workflow that just clicked.
Two Kinds of Memory
What We Learned Building an AI That Remembers I spent an evening this week untangling a problem that sounds small and turned out to matter a great deal. We were building memory into the AI system that runs our agency, and I kept getting it wrong in the same way. The fix taught me something that reaches well beyond software, so it felt worth sharing. Here is the heart of it. When you ask something to remember, you are really asking for two different things, and they want opposite treatment. The first we call the learning log. It holds what you have decided and how you work right now. It is alive. You update it when something changes and you take things out when they are resolved. It is meant to be tidied, because its whole job is to reflect the present. The second we call the activity log. It holds what actually happened, and when. A piece of work delivered. A sign-off given. A decision made on a Tuesday. You only ever add to it, and you never rewrite it. Its value comes from staying untouched, because that is what makes it trustworthy later. So one is constantly revised and one is never revised. That single difference is the whole story. Why keeping them apart matters: The mistake we kept making was pouring both into one place. Do that and neither job gets done. If the daily record floods into the "how we work now" file, that file stops being a clear picture of the present. You end up wading through a diary to find the one rule you actually need. And if you keep your evolving decisions inside the daily record, you can never tidy them, because the record is supposed to stay frozen. Two honest jobs, pulling in opposite directions, sharing one drawer. The part that surprised us The harder question was not where to keep things. It was what is even worth writing down. Our first instinct was to log everything, every small step the system took. That turned out to be noise. The rule that actually worked was simple: only record what a person would genuinely mention in a monthly review. Something delivered. A decision reached. A problem raised or cleared. Everything else, the busywork in between, does not earn a permanent line. A memory that records too much is almost as useless as one that records nothing, because you can no longer find the moments that mattered.
To Claude or not to claude
I've been using Claude exclusively for a few months now. Started with openclaw, but that's a whole other story... I'm considering switching to antigravity. Gonna test it out tonight but wanted to see what you guys think about it as far as reliability and coding, etc Any direction is appreciated.
1-30 of 2,168
Clief Notes
skool.com/cliefnotes
What we give away free beats most paid courses. Build durable AI systems with a Marine vet and Edinburgh researcher. 40+ lessons, growing.
Leaderboard (30-day)
Powered by