AIMM Hub

Mastermind recap

AIMM Session — September 18, 2025: NHS Voice Agents, GEO Without Backlinks, and the Sourdough Excavator

· AIMM 2025 · 90 min

Facilitators: Lou D'Alo

“Let AI commoditize the average. We’ll work on the extremes.” — Lou

30-Second Summary

This pre-holiday session packed more practical firepower than most full-day workshops. The group swapped real experiments with Claude Code and vibe coding, surfaced a breakthrough insight on AI memory and knowledge management, dove into a real-world NHS voice agent case study that cut patient wait times by 71%, and then Lou dropped what may be the single biggest GEO opportunity any of us have seen — a Python-generated, schema.org-powered FAQ page that could get you cited by ChatGPT, Perplexity, and Grok without writing a single blog post. Plus: a Socratic chatbot to help you excavate what AI can’t replace about you.

1. Vibe Coding Is Real — But the Learning Curve Bites

Kasimir opened with a clean win: using Claude Code (the command-line interface from Anthropic), he went from zero to a working interface for a Pinecone experiment in under an hour. The tool fetches its own documentation, self-troubleshoots, and asks for approval before executing. Hands-off, mostly.

Lou layered in the honest caveat: it’s not the “Twitter clone in 5 minutes” hype you see on YouTube. The real workflow is:

  1. Talk to the AI until your requirements are right
  2. Ask it to generate a requirements doc
  3. Feed that doc to the coder
  4. Implement one thing at a time

Skipping that process — as Lou demonstrated when a live demo collapsed over a misidentified API authentication method — leads to rabbit holes that eat your day.

Lou also mentioned an open-source orchestrator that lets you run 5–6 Claude Code sessions concurrently on the same codebase without them overwriting each other.

2. The Knowledge Management Gap Nobody’s Solved Yet

Don named the problem everyone feels but rarely articulates: ChatGPT has built up a rich history of everything he’s worked on — but he can’t find a damn thing in it.

Elizabeth brought the gem: Perplexity’s artifact view lets you save outputs directly to Google Drive (or download as .md) without copy-pasting. She also used Perplexity’s Deep Research mode to find updated versions of the Infinite Prompt Generator, compare old vs. new, and synthesize a hybrid — then turned it into a new Claude skill.

Lou’s proposed pipeline: Perplexity artifact → Google Drive → N8N trigger (3 simple modules) → Qdrant vector store.

Kasimir added: Claude’s memory feature (opt-in under Settings → Personalization) now lets you reference previous chats directly. No more handover documents.

3. Voice Agents in Healthcare: The DORA Case Study

Bally shared a visit to a UK university project that deployed an AI voice agent called DORA for post-cataract surgery follow-up at an NHS trust.

Results:

  • 35 weeks → 10 weeks patient follow-up cycle time
  • 92% patient satisfaction — from an elderly population who knew they were talking to an AI
  • Built by a startup using WAPI, with applied linguistics researchers from Oxford and Newcastle shaping the conversational design

The key engineering wasn’t the AI — it was the linguistics. Researchers tracked interruptions, latency, filler words, and cultural communication patterns.

Don’s context: he spent years as VP of Health Innovation for an Alberta agency. The root issue in healthcare isn’t funding — it’s throughput. Voice agents win because staff resist outbound calls (repetitive, draining), bots never fatigue, patients can’t leave a message (no dead-ends), and the system is multilingual without misinterpretation risk.

This was the main event. Lou unveiled what he called “the single most powerful authority play we’ve ever had in the mastermind” — a collaboration with Ken Droz (of the Ideal Client Handbook platform) and a long conversation with Grok.

The insight: Old SEO is dead. The new game is showing up in generative engines. The mechanism: the knowledge graph, specifically schema.org structured data embedded in a single HTML page.

What the tool does:

  • Takes your Ideal Client Handbook (ICH) as input
  • Extracts every fear, belief, value, desire, and identity statement
  • Generates ~400 long-tail questions your exact client is typing at 3 AM
  • Answers each one in your voice, subtly positioning you as the solution
  • Embeds entity signals linking to Wikidata and causal relationship chains
  • Wraps it all in schema.org markup

The result: you paste one code block into a single HTML page on your site. No visible content required. No backlinks needed. The machines do the indexing.

Projected outcomes (per Grok’s estimates): In Google’s knowledge graph within 4–5 days / 5,000–50,000 impressions/month within 3–6 months / 200–2,000% spike in unbranded search traffic / 5–30+ AI citations per month in Perplexity/ChatGPT/Grok.

Lou ran the Python script (geox1.py) live, generated the output for his own ICH, and confirmed the structure looked correct. Not live yet — he’s packaging it into something easier to deploy.

Important: Ken Droz and Eben Pagan may productize this. Use it quietly for yourselves for now.

5. The Sourdough Excavator: Monetizing What AI Can’t Replicate

Core idea: “Monetize Your Bias.” The analogy: sourdough bread. What makes sourdough more valuable than commercial bread is the introduction of a contaminant — something that doesn’t belong in the ecosystem — that creates a new and valuable product. Your unusual element, your weird perspective, your hard-won scar tissue — that’s your sourdough starter.

The Flash Boys parallel (hat tip to Don): Mike Novogratz made 8 figures annually arbitraging the spread between buy and sell. Then algorithms got faster, and the spread collapsed to nothing. The information didn’t disappear — the gap did.

Lou’s reframe: AI converges expertise to the average. The value is at the edges. What AI can’t yet do: tell a client “that won’t work — I’ve tried it” / apply pattern recognition from 20 years of watching launches fail / know what not to do.

Lou’s business model vision: give away everything that’s commodity (frameworks, templates, FAQs) at low or zero cost. Charge premium for judgment, experience, and the coaching that prevents expensive mistakes.

Hot Takes

  • “It’s easier to vibe code than it is to wire up an N8N flow anymore. I don’t even remember how to use N8N inputs and outputs.” — Lou. The abstraction layers are collapsing.
  • “Humans have a high reluctance to pick up the phone and do a cold call. A bot is ideal for that.” — Don. Healthcare AI wins not because bots are better than humans — but because humans hate repetitive outbound work.
  • “Knowing what not to do is just as valuable as knowing what to do. I don’t think AI is in a position to do that yet.” — Lou. This is the positioning frame for every knowledge entrepreneur in this room.

Try This Before Next Session

Run the Sourdough Test on yourself. Open a new Claude or ChatGPT conversation:

“I want you to act as a Socratic interviewer helping me find my differentiated positioning as a knowledge entrepreneur. Ask me one question at a time about how I see problems differently than my peers, what mistakes I’ve watched others make that I’ve avoided, and what I know that isn’t obvious even to other experts in my field. After 5–6 exchanges, reflect back what seems genuinely unusual or irreplaceable about my perspective. Don’t summarize — push me to be more specific.”

Then bring your answers to the next session.