THE AI UPSKILL
A weekly digest for Beginner Builders
⚡ TLDR
This week we're showing you how to build an AI lead qualifier that scores every inbound lead automatically, logs it to a spreadsheet, and pings your inbox when a hot one lands. No coding required.
We've sifted through the latest YouTube tutorials and picked out two that are ideal for Beginners, on LLMs and Claude Cowork. In other news, Claude gets into a spat with the Pentagon and takes top spot for most downloaded LLMs, while ChatGPT controversially starts running ads for its free tier.
ARTIFICIAL INTELLIGENCE
THIS WEEK WE'RE WATCHING
→ A beginner’s guide to using LLMs effectively: This is ideal for the absolute beginner or those who want a quick reminder on the tips and tricks to get the most out of ChatGPT and AI more generally. Ali Abdaal covers prompts, efficiency hacks and more. Well explained an approachable. Watch here.
→ How to use Skills in Claude Cowork: Mastering skills, whether in Claude Code or Cowork, is a huge power unlock. Using skills, Brock Mesarich is able to turn a YouTube video into Linked In content that is automatically posted by Claude. But this video covers so much more, including how to set up your Claude for maximum efficiency through skills, connectors and more. Watch here.
UPSKILL FURTHER
WANT TO GET MORE OUT OF THIS?
Want to go deeper on any of this? Join the AI-Enabled Marketer Skool community, which takes you from absolute beginner to AI-Enabled Marketer with Myosin.xyz, a community of top marketing professionals.
Join the movement here - https://www.skool.com/the-ai-enabled-marketer/classroom
THE BUILDERS TIP
🛠 SIMPLE BUILD: AI LEAD QUALIFIER
Most leads sit in a spreadsheet until someone gets around to reading them. By then, the hot ones have gone cold.
This week's build is an AI-powered lead scoring system.
Someone submits your form → the AI reads their budget range, what they need, and how they found you → it scores the lead using criteria you define (budget weight, request specificity, referral bonus) → categorizes them as Hot, Warm, or Cold → logs everything to your tracking sheet → and pings your inbox if it's a Hot lead so you can follow up fast.
QUICK START:
Download the workflow JSON and import it into n8n
Create a Google Sheet called "Lead Tracker" with these columns: Name, Email, Company, Budget, Request, Source, AI Score, Category, Priority Action, AI Reasoning, Date (Make sure to open access to the sheet)
Connect your credentials:
OpenAI — on the AI Lead Scorer node
Google Sheets — on the Add Lead to Sheet node. Select your Lead Tracker sheet from the dropdown list, then map each column value: set Name to { $json.Name }}, Email to { $json.Email }}, and so on for every column. Each value comes from the Parse AI Score node output.
Gmail — on the Alert: Hot Lead! node
Open the Alert: Hot Lead! node → change the "To" email to your actual email address. Optional: under Options, turn off Append n8n Attribution for cleaner alert emails.
Click "Test Workflow" at the bottom of the canvas — this will open your lead form in a new tab
Test it: Fill out the form with a fake high-value lead — pick "$10,000+" budget, write a specific request like "Need an AI chatbot for our customer support team handling 500 tickets/day," and select "Referral" as the source. Check your Google Sheet — you should see a row with a score of 9 or 10 and a "Hot" category. Check your inbox — you should have an alert email.
Now test a cold lead: "$1,000" budget, vague request like "just exploring options," source "Other." It should score 2-3, categorize as Cold, and — importantly — NOT send you an email.
Something not working? Copy the Parse AI Score output and paste it into Claude or ChatGPT for troubleshooting.
MAKE IT YOURS:
Change the scoring criteria — The AI system prompt spells out exactly how it scores. Want to weigh industry or company size more heavily? Edit the scoring rules in the AI Lead Scorer node's system prompt. The AI follows whatever logic you give it.
Add more form fields — Need to know their industry? Timeline? Tech stack? Add fields to the Lead Form node and update the AI Lead Scorer prompt to factor them into the score.
Customize your alerts — Right now, only Hot leads trigger an email. Want Warm leads too? Duplicate the IF node and Gmail node, or change the IF condition to check for score >= 5 instead of category = "Hot."
Swap the model — We use gpt-4o-mini for speed and cost. For more nuanced scoring with longer requests, try gpt-4o or Claude Sonnet.
Connect it to your real website — The form trigger generates a URL you can embed or link to. Or replace the Form Trigger with a Webhook node and connect it to Typeform, Tally, or your existing website form.
To access the JSON you need for this build, just head on over to the AI Enabled Marketer Skool community through the link here - remember your first 7 days are free!
The Battle of the LLMs heats up
Claude hit #1 on the App Store after the Pentagon drama and is now outperforming ChatGPT in blind writing tests. OpenAI responded by rolling out ads on its free tier, something Anthropic immediately positioned against. Meanwhile, both Google and OpenAI dropped significant model updates, and the emerging consensus is clear: there's no single best AI anymore. Use the right tool for the right job.
One Startup's Answer to AI Hallucinations: Ask 10 Models at Once
CollectivIQ queries ChatGPT, Gemini, Claude, Grok, and up to 10 other models simultaneously, fusing their answers to produce more reliable results than any single model alone. This is your periodic reminder that LLMs can hallucinate and so, when something matters, try the same prompt in Claude, ChatGPT, and Gemini and compare what comes back.
That's it for this week.
Building something cool with AI? Reply to this email. I feature reader projects. 💬
P.S. If this was useful, forward it to someone upskilling in AI.