User
Write something
🔒 Q&A w/ Nate is happening in 8 minutes
Pinned
🚀New Video: Every Claude 2.0 Concept Explained for Non-Coders
In the last six weeks Anthropic shipped three new models, put docs, slides, and designs inside every chat, launched a new agent app called Duet, and dropped the 5-hour limit on the $200 plan, so people are calling it Claude 2.0. In this video I break down 25 Claude concepts in plain English across five parts: the models it runs on, where you use it, what it can make for you, what it knows about you, and how it works on its own. You'll learn tokens, effort levels, plans, Cowork, artifacts, memory, skills, connectors, mods, sub-agents, /goal, and more, and you don't need to code for any of it. If you only set up three things today, start with a claude.md, one skill, and a few /goal prompts. CLAUDE 2.0 CHEAT SHEET
Pinned
ANNOUNCING: A certification for AI builders who want bigger clients
Building with AI gets easier every month. Landing the bigger clients doesn't. Because the moment you pitch a growing company, the questions change. - Legal asks how you're handling data governance. - Finance asks what it costs to run at scale. - Procurement asks how they know it'll work reliably. Most builders don't have those answers. And that's where the deal dies. So we built a program to fix it. -> ANNOUNCING: A certification for AI builders who want bigger clients It's called the Certified AI Consultant program. Over 10 weeks, it teaches the 7 skills bigger companies actually pay for: scoping, pricing, governance, evals, AI economics, and more. Then you earn the credential through a graded portfolio and a live interview with my team. It's already working. Reidar, from our first cohort, closed a $75,000 contract with a global engineering company using what he learned. Cohort 2 starts November 2nd, and it's capped at 50 seats. Let me know if you have any questions. Talk soon, Nate
Pinned
What do you get if you upgrade to AIS+?
Some of you have never heard of the AIS+ community. Others have but the part that trips you up is the actual difference between the two. Either way, this post will give you clarity. This free group is a bundle of quick resources pulled from my YouTube videos, plus a massive open community that anyone can join. It's a great place to get your bearings and see what's possible. But it's open to everyone, it can be noisy and overwhelming, and there's no path through it. You can get help from other members, but I rarely answer questions here. AIS+ is the opposite: - A step by step roadmap with a clear order, so you're never guessing what to do next - A much smaller community of people who are seriously committed to building and selling AI agents - I answer questions every day and run a weekly Q&A call where you can get direct access to me For the course material: The roadmap takes you from zero to building and selling AI agents, and the whole thing is built on the latest tech like Claude Code and Codex. We update it constantly. The old n8n material has been archived. It's still there if you want it, but it's no longer the focus, because the way you build today has moved on and the courses moved with it. Here's the actual roadmap inside, in order, with when each piece opens up: 1. Start Here (opens the moment you join). Gets you oriented. How the community works, the path ahead, and how to get help when you need it. 2. Build Your Portfolio (opens the moment you join). Why a portfolio matters, beginner level tutorials, and what types of projects to focus on. You end up with real work you can show a client. 3. Claude Code (opens the moment you join). This is now its own dedicated course. Build faster, turn ideas into working automations, and go deep on the tool serious builders are using right now. This takes you from beginner to advanced, step-by-step. 4. Get Your First Clients (opens after 30 days). Getting your first clients is hard, because you don’t have any case studies yet. So, we analyzed all of the success stories from our members and found they get their initial clients with two different techniques: warm outreach and Upwork. So, we teach both techniques in detail with exactly what to say, exactly how to position yourself when you have no proof.
A dictation benchmark hid an 86% vs 37% gap
My local 4B model looked equal to Groq on an initial comparison of three real dictations. Those cases contained only two fillers total. There wasn't much there to test filler removal. I repeated the comparison on 6 filler-heavy real dictations containing 49 fillers, through FluidVoice's live loopback. Groq gpt-oss-120b with medium reasoning removed 86% of fillers. Local Qwen3.5-4B removed 37%. On 2 of the 6 dictations, the local model stripped zero fillers. That changed the recommendation: Groq remained the default, and local inference became the offline fallback. Network failures affected 4 of 233 rewrites, so there's still a practical role for the local model. The misleading part was the initial tie. The cases were real, but they barely contained the thing I was trying to measure. For cleaning dictation, I'll benchmark on filler-dense real cases so the comparison can actually distinguish the models. Have you had a benchmark suggest two models were equal until you tested harder real examples?
I raced the director pattern against one plain AI on a real job. The plain AI won, except for one part
Every week a new AI workflow goes viral. Most of the time the setup you already use holds up once you test both on real work. The director pattern: the AI you talk to stops doing the work, splits the job into pieces, and hands each one to a helper AI with a short rule sheet. When something breaks, you fix one rule sheet instead of starting over. On October 26 I tested it on a real job for my wife's online pet-toy shop: two weeks of holiday Instagram posts and Pinterest pins, six of each, real product photos only, every draft checked against her claim rules before her approval. Director with five helpers vs one plain AI, same prompt: - The director cost 48% more ($8.17 vs $5.52) and ran 12 helper tasks across three rounds of fixes. - The plain AI did better creative work. It read camera data on the image files to find six real iPhone photos and laid out 26 ready-to-view slides. The director grabbed website cut-outs and 640-pixel images. Two parts did earn their keep: 1. A separate rule-checker with a numbered pass-or-fail list failed five of six drafts in round one. It caught a photo file named after an AI prompt, safety-sounding lines, and toy sizes printed on photos that the plain AI missed. 2. When I asked mid-job to turn one post into a Black Friday store reminder, it changed only that item in under 10 minutes and flagged that it broke the shop's no-store-campaign rule for late October through December. A separate check is also why we review client builds before launch. When I audited an app a founder built with a Replit agent, it worked on the surface, and the bugs that mattered sat underneath: a master database key pasted into a config file, and two-factor login checked only on the phone screen, not the server. How I test any hyped method now: 1. Have your agent compare the new tool against your current setup before switching. 2. Race both on one real job. 3. Keep only the part that beats your baseline. If you want the launch-readiness checklist I run on AI-built apps before they go live, comment below and I will share it.
1-30 of 26,690
AI Automation Society
skool.com/ai-automation-society
Learn to get paid for AI solutions, regardless of your background.
Leaderboard (30-day)
Powered by