Activity
Mon
Wed
Fri
Sun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
What is this?
Less
More

Memberships

Constellations

80 members • Free

Clief Notes

35.8k members • Free

37 contributions to Clief Notes
🏆 WEEKLY COMP #6: THE RESEARCHER 🏆
🎟️ PRIZE: FREE SEAT IN THE LYCEUM 🎟️ Pick your cohort. Technical, Business, or Creator. Your call. ---- 🇺🇸 Quick note first. This post is going up Today because we took Memorial Day off yesterday. To keep things fair, you've got until Sunday May 31st at 12:00 PM EST to submit. Same week of build time, just shifted. ---- 📋 THE CHALLENGE Build a folder-based AI researcher for a specific topic or industry. You pick the domain. This week's deliverable is one researcher folder that someone could drop into a Claude project and use as their personal research partner for whatever domain you've built it for. ---- 🎯 PICK YOUR DOMAIN The domain is yours. Pick something specific. Pick something you'd actually use. A few sparks to get you thinking: - 🏦 M&A activity in one industry (fintech, healthcare, defense) - ⚖️ Court cases in one area of law (employment, IP, immigration) - 🧬 Scientific research on one health condition or treatment - 🏘️ Real estate market dynamics in one city or asset class - 🥊 Competitive intelligence for one product category - 📜 Historical research on one period, place, or movement - 📚 Academic literature in one specific subfield - 📋 Regulatory developments in one sector - 📰 Journalism research on one beat (climate tech, AI policy, biotech funding) The more specific, the better. "Research assistant" is too broad. "M&A research analyst for early-stage fintech deals in the US and Europe" is right. ---- 🗂️ THE METHODOLOGY If this is your first comp, welcome. Here's what you need to know: This week (and every week) you're learning interpretable context methodology. Folders as architecture. Each file does one job well. Your researcher is a folder with five things: - 📄 identity.md (who the researcher is, what domain they cover) - 📐 rules.md (how they research) - 💬 examples.md (what good looks like) - 📚 reference/ (frameworks, source lists, key concepts) - 📖 README.md (how to use it)
9 likes • 2d
🏆 SUBMISSION 🏆 Getting Started: https://usedcolouringbook.github.io/aeo-geo-researcher/ Repo: https://github.com/usedcolouringbook/aeo-geo-researcher/ My Researcher for this week came from a problem I've been trying to answer: how do we design content to let Search 2.0 do marketing for us? The whole point of ads & marketing is to get your product to people who are willing to purchase it. The Agentic Search landscape (Search 2.0) is about finding the best answer to your question. To me, the current landscape is almost primed to create an environment where if you make your content attractive and easy to verify for agents, then they'll naturally push your content towards everyone who asks about anything related to it; to borrow the meme: "It's free real estate". The AEO/GEO Researcher is your assistant and toolkit so a new web developer & the savvy B2B SaaS owner can both find out where and how they can improve their content for agentic use, which platforms are impacted the most by issues/resolutions, and where to go to keep learning after you've closed the session. There are 2 different entry points for this tool: Claude.ai & Claude Code. Each one will work as an assistant, with Claude.ai usining a Claude Project in order to take a "snapshot" of the url and generate a research report after a session. You can start a new chat by feeding it the research brief from the previous session in order to build upon previous work but without the actual context. The Claude Code version is more robust version than Claude Projects, with session persistence, live queries, data enrichment, and a tool that can implement the fixes instead of just writing a gap analysis. One thing I'd like to note is that the WebFetch issue I uncovered a few weeks ago has been natively baked into the auditing process and has in-depth explanations for the workaround of using 'llms.txt' & 'llms-full.txt' to pass along structured information about the site that gets stripped from WebFetch and other agent HTML-to-MD scraping tools.
Claude completely ignoring folder structure this week
This week, my token usage jumped, and the quality of responses plummeted. Then, I started asking directly if it read and/or followed the folder's Cluade.md and Context.md. I've attached the response. Anyone else having this problem? —edit (IMPORTANT)— Impossible to know for sure, but my research/testing over the weekend strongly suggests an update to Claude desktop last week likely raised the temperature at which Co-Work sends your requests to the Claude API, so it gets loosely-goosey with your folder structure’s governance (to the point of ignoring most of it, regardless of model, regardless of how strict the wording). If you need more deterministic/ governance-driven usage of your folder structure (=more consistent, predictable outcomes), use Claude Code, whose temperature set to 0.0 by default for maximally deterministic coding. If you want more creative/interpretive/random outcomes, go with Co-work.
Claude completely ignoring folder structure this week
2 likes • 2d
This happens to me all the time. I've got hooks set up to load 3 documents and it's never worked. I have to tell claude to read it's SOUL doc each time I start a new chat because the hooks & memories don't surface until under pressure or close to my usage limit. I also have a hard time getting it to use scripts with explicitly telling it when to use them, even though the document it's reading from has an actionable table for scripts, which are used in the document to accomplish the work. Some days though it follows everything and it's AWESOME for that small time frame haha.
🚨 You've been asking when the Lyceum opens. The waitlist is live. 🚨
The waitlist is up and seats are limited, so this is your nudge to go lock yours in. 👇 New here? Quick context. 👀 The Lyceum is Jake's live cohort program built on ICM, the methodology 35,000 people in this community are already using to get real results with AI. The short version: folders over agents. You learn the layer underneath the tools, the one that keeps working when the next model drops. Full breakdown is on the site. Here's what's inside: 🎯 Three cohorts, Technical, Business, and Creator. Same methodology, built around what you actually do. 🎥 Live sessions with Jake and a full team of instructors. ♾️ Lifetime recordings, written curriculum, and a private cohort Discord. 📜 An Eduba ICM certification you can put on your resume. And a guarantee no course makes: ✅ You leave with a working product, or the team finishes it with you. ⏳ Seats are limited and this community moves fast, so the math is not in your favor if you wait. 💡 Pricing and start dates aren't public yet. The waitlist sees them first, gives feedback on timing, and gets in before the program opens. Everything you want to know is on the page. If you already know this is for you, get on it. 🔥 👉 https://lyceum.eduba.io
2 likes • 3d
FINALLY!
Who's here? Drop your intro.
Tell us three things: 1. What you do (job, industry, student, career-changer, whatever) 2. What brought you to Clief Notes 3. One thing you're trying to figure out right now related to computing or AI I'll respond to every single one. And read each other's intros too because the person who's stuck on the same problem as you might already be in this thread. I'll go first I am Jake, I have been working in tech for 15 Years, building with Generative AI for 3 Years straight now! Excited to teach and learn! That's it. Simple, scannable, gives you data on who's joining and what they need, and keeps the feed clear for content that retains people past week one.
3 likes • Mar 14
@Isaac Safdeye Check out the first Afternoon Tea with Joe Fioramonti, he's got a great application up and running that may be in line with what you're looking to do. Hope this helps.
1 like • 8d
@Marc Nicholson I just came across this post a few minutes ago and I think it may be something you could use and apply right now: www.skool.com/cliefnotes/free-gold
Free gold
I just compared this setup to the winner of the anthropic hackathon with qwen 3.7 max - it came out on top by a fair few percent though mine is project specific not generic, that said here you go a generalized prompt to set up an operational workflow where the human matters - this works for me day in day out and was rated higher that a $146k hackathon winning set up Love to hear feedback or critique - love even more to hear how this is working great for you. Its development focused, it expects you know what you are doing, but only to the point of designing and then testing after the fact - I am a tester by trade with solid infra/sec understanding. Real life usage of this implements a bunch of different skills, a very small set of agents and I am playing around with use of sub agents at the moment.
3 likes • 8d
Holy cow this awesome! I think a document like this is what I've been missing from the kit, I have all the separate pieces but nothing to link them together. Very excited to see where this goes!
1-10 of 37
Crae Säkkinen
4
89points to level up
@john-doe-4388
Just vibing through

Active 10h ago
Joined Mar 11, 2026
INTP
Powered by