25 years in marketing, and I’ve never been this curious about what’s possible. Exploring what AI lets us do. Sharing what I learn.
Vancouver
Joined July 2026
- Tweets821
- Following84
- Followers17
- Likes100
Pinned Tweet
AI doesn't save you time on day one. It saves you time every week after.
Here's what that looked like when I built the agents behind my newsletter, and how I'd start if I did it again.
Derek Reynolds reports HighLevel’s Ask AI and AI Studio are free under fair use. A free seat isn’t a free workflow. My Cost Desk test: time a client campaign brief, draft to approval, with and without AI. See how much of that process AI takes off our hands. #marketingmultiplied
Someone searching “best aeo agency” isn’t after another explainer. They want to see what they’d buy. Scott Pang’s sample caught that query. I’d hand Claude my real scope, inputs and deliverables and have it build that page. Wild what AI can package now. #marketingmultiplied
Matt Pocock's prompt is one question: "How do we make this change fully verifiable?" I'd steal it for marketing ops. Before a page, ad or GBP update ships, decide what proof will show it worked. Then AI helps you build the proof, not just the change.
#marketingmultiplied
Greg Isenberg's /grill-me is stealable: interrogate the idea before anything gets built.
I'd write my call, then let an agent stress-test assumptions. Explore→exploit before you build. One person can pressure-test a day of work first.
#marketingmultiplied
running a company nowadays feels kinda dystopian
my team is 50+ people in 12 countries, but i talk to my agents more than my human teammates
- voice note to grokbot
- mattpocock/skills /grill-me so the agent interrogates the idea before anything gets built
- spec drops into linear through the linear mcp
- exa + firecrawl agent reads every competitor site, pricing page and changelog
- ideabrowser.com to find ideas and trends
- apify pulls g2, reddit and app store reviews
- prompt: "find the 10 complaints that show up most and quote the users"
- context7 so the agent reads the latest docs before it touches any api
the design
- claude code + paper mcp
- 4 versions of every screen before a single line of code
- pick pieces from 2, it merges them. diverge then converge
- frontend-design skill from anthropic/skills + our own design system skill
- cleanshot screenshots of any ui i like go straight in with "make it feel like this"
the build
- pingdotgg/t3code as the control surface for every agent, from my phone too
- stablyai/orca runs a fleet of agents in parallel, each in its own git worktree
- graphify maps the whole codebase into a knowledge graph so agents navigate instead of guess
- headroom compresses tool outputs before they hit the model, so tokens go 20-95% further
- lsp plugins so the agent sees types and errors like an ide does
- security guidance hook flags risky edits before they land
- supabase mcp for the database, stripe for billing
- github mcp opens the pr, vercel spins up a preview
- loops run overnight, i check what shipped over coffee
the user testing
- lightpanda as the headless browser, way lighter than chromium so you can run hundreds at once
- agents click every flow on the preview like a first-time user
- one persona is impatient, one is confused, one is on a phone
- playwright + chrome devtools mcp for regression tests
the bugs
- sentry mcp catches the error
- agent reads the stack trace and opens a fix pr in under 10 minutes
- human just approves
the data
- posthog mcp cron at 7am on yesterday's funnel
- any step that drops more than 10%, the agent watches 20 session replays and tells me why
- stripe mcp pulls every cancellation reason into one doc every monday
the agents' tools
- orgo so every agent gets its own computer
- agentmail so every agent gets its own inbox
- agentphone so i can call them from the car
- composio connect-apps for every other connector
- akitaonrails/ai-memory, a git-backed markdown wiki so they remember what we decided last week
the skills
- wshobson/agents so skills work across claude code, codex and opencode
- my writing voice
- our design system
- how we write specs
- skill creator so the agents write new skills themselves
some helpful tricks
- one agent per platform. the x agent never touches email
- cheap model for triage, opus for anything that ships
- every output gets scored 1-10 by an llm judge. under 8 goes back
- every agent writes a 5-line log at night so i can read the whole company in 2 minutes
- every agent mistake becomes a new skill, so it only happens once
the team
- humans get version 5, not my voice note which they are happy about lol
- their job is creative feedback, relationships and telling the agents what they got wrong
a human pm reviews the funnel once a week. this reviews it every morning
running a company is now voice notes and inference
OriginAI maps an AI agent team with a leader bot. What I'd do first: skip the leader. Lead 2-3 narrow bots for a few weeks. Save the guidance you keep repeating as the leader's instructions. One-person creative team via Explore→exploit.
#marketingmultiplied
🚨 You're wasting Claude's potential by treating it like a single AI assistant.
What if you had an entire AI team working together instead? 🤯
Here's how to build your first team of AI agents in Claude 👇
🧠 1. Team Leader
Breaks down goals, delegates tasks and combines results.
🔎 2. Research Agent
Finds information, analyzes data and checks sources.
✍️ 3. Content Agent
Creates X posts, LinkedIn content, hooks and drafts.
📈 4. Marketing Agent
Plans campaigns, positioning and audience targeting.
🛡️ 5. Quality Agent
Reviews accuracy, clarity and content quality.
⚡ How the workflow works:
Goal → Delegate → Research → Create → Review → Final Output
Start with just 3 agents:
• Research
• Content
• Quality
Use Claude Code + Agent SDK for programmable agent workflows or start simpler with Claude Projects + Skills.
Don't build a complicated system on Day 1. Make one workflow work reliably, then expand your team.
Save this guide. ♻️ Repost it.
Follow @Origin_AI_01 for practical AI workflows.
#ClaudeAI #AIAgents #ArtificialIntelligence #Automation #AITools #FutureOfWork
Lily Ray sees GBP embeds increasingly cited in AI Overviews and AI Mode. Your Business Profile is a second homepage AI may quote.
I'd write services, hours and towns as citable facts AI can attribute. One careful pass can change what AI says about you.
#marketingmultiplied
Another day, another Google SERP feature that is exploding in organic visibility and AI Mode/AI Overview citations.
This time google.com/searchviewer/ - a subfolder containing Google Business Profile embeds for businesses that's increasingly cited in AI Overviews and AI Mode.
So now websites get to compete with Google Shopping results, Google Knowledge Panel results, Google flight URLs (h/t @natzir9), Google travel URLs, and probably other things I'm forgetting - directly in those coveted AI citations.
1/ Your conversion report needs a new question: how much of this activity involved an AI agent? If you’re paying for traffic, you need to understand what reached the checkout or booking form before trusting the total.
6/ Observe, try, fix, repeat. We learn where agents get stuck and check what our conversion numbers actually mean. Each fix helps the customer path. Each check helps us judge the traffic.
#marketingmultiplied
7/ Source: ppc.land/checkout-appears-in…
1/ A blog full of useful posts can still leave you with an image backlog. @chris_nectiv shared a way to turn Screaming Frog into a bulk image studio. He reports 100+ blog images in 10 minutes. One crawl, one batch, a whole backlog moving at once.
6/ What I’d add: one house-style line covering brand colours, no text in the image and no fake people. Then I’d spot-check before publishing. The batch saves us production time. The shared style makes the images look like they belong together.
#marketingmultiplied
7/ Source: teapawt.net/chris_nectiv/status/21…
SEO Tip: You can leverage Screaming Frog and the Nano Banana API to create blog images at scale. In 10 minutes, you can generate 100+ blog images:
So this is a pretty nice use case of AI in my opinion that can help with asset creation. A lot of people don't know that you can hook up @screamingfrog to basically an LLM API that you want to. It can be really nice to get some AI integrations directly out of the tool in a platform that many SEOs are familiar with.
Well one thing you can do is leverage it to create blog images, especially if you don't have the internal resourcing to design assets with every single blog post. Nano Banana is pretty darn good at image creation. Haven't tested all of the new Gemini updates quite yet but imagery is where I think the model shines.
Here's the process:
1. Open up Screaming Frog
2. Navigate to Configuration > Spider > Extraction. Turn on "Store HTML and "Store Render HTML"
3. Now go to Configuration > API Access > AI > Gemini. Use your API key to connect to Gemini
4. Click "Prompt Configuration". Select the model category as "Image Creation" and model name as "gemini-nano-banana-2.1".
5. Configure your prompt. I feed it the entire Page Text and then have it create a "flat vector illustration"
6. Start your crawl. In the "AI" tab on the right-hand side, you'll see your images appear. You can also export them all in the export settings.
So a pretty nice process to get image assets created in a common tool that marketers all use.