The Founder's AI Stack: 7 Tools Running My Business in 2026

Last Tuesday at 6:40 AM I sat down at my desk, opened my laptop, and realized something strange. My business had already been working for two hours.

A Make scenario had pulled overnight client check-in data from Firebase into a Notion dashboard. Claude had drafted three personalized coaching follow-ups based on patterns it flagged in those check-ins. My scheduling tool had confirmed two diagnostic calls for the afternoon and sent both prospects a prep questionnaire. And an analytics summary was sitting in my inbox showing me which content from last week actually moved the needle.

I hadn't touched any of it. I picked up my coffee and started my actual work — the strategic thinking, the creative direction, the identity coaching that no machine can do.

Two years ago this operation required five people. Now it's me, one senior strategist, and a part-time editor. Not because AI is some miracle. But because I rebuilt everything around a simple question: what does AI do well, and what do humans do well?

The Question Nobody's Asking

Every founder I coach asks me what tools to use. They want the list. The stack. The setup.

Wrong question.

The right question is: what kind of founder are you trying to be? Because your AI stack is a mirror. It reflects your identity back to you — specifically, whether you're still operating as an Operator or an Architect.

I had a founder come to me last fall — call him Marcus — running a $2M SaaS company. He'd bought every AI tool on the market. Literally seventeen subscriptions. He was using AI to answer customer tickets faster, write more Slack messages, generate more meeting agendas, and process more support requests per hour. He was doing more, faster, all day long.

He was also burning out.

Marcus had built an Operator's AI stack. Every tool was designed to help him do more of what he was already doing. The AI was amplifying his workload instead of replacing it. More throughput. Same bottleneck. He was the fastest hamster on the fastest wheel.

When I rebuilt Marcus's stack from seventeen tools down to six, his company grew 40% in the next quarter. Not because the tools were better. Because his time was finally going to the right places.

What My Week Actually Looks Like

I run three things: a coaching practice for tech founders, a dev studio that's shipped 600+ projects, and a content operation that feeds both. Here's how AI shows up in an actual week — not in theory, but on my desk, Monday through Friday.

The Brain: Claude

This is where my strategic thinking happens. Not "happens" like I outsource it — "happens" like I think with it.

Monday mornings I do a founder review for every active coaching client. I paste their weekly check-in data, their goals from our last session, and any notes I've taken during the week. Then Claude and I have what I can only describe as a conversation with someone who has perfect memory and no ego. It'll catch patterns I miss — "this is the third week this founder has mentioned team friction without naming the specific person" — and I'll catch context it can't. I know that founder just had a baby. I know their co-founder is thinking about leaving. I know the tone beneath the words.

That combination — machine pattern recognition plus human contextual judgment — is where the real value lives. Neither one alone gets there.

I also use Claude for first drafts of everything. Blog posts, email sequences, workshop outlines, client frameworks. The first draft is maybe 60% of the way to done. But that 60% used to take me four hours. Now it takes twenty minutes. And my editor — a human with taste and voice intuition — brings it the rest of the way.

The Nervous System: Make and Zapier

If Claude is the brain, Make and Zapier are the nervous system. They're the connective tissue between every tool I use.

Here's a specific example. When a founder books a diagnostic call through Cal.com, a Zapier automation immediately creates their profile in my Notion CRM, sends them a welcome email with a prep questionnaire, notifies me in Slack with a summary of who they are and where they found me, and — if they came from a specific blog post — tags that traffic source so I can see what content is actually driving calls.

That used to be a VA's job. Four steps, fifteen minutes per booking, twenty times a month. Five hours of work that now happens in three seconds with zero errors.

But here's the part most founders miss: the automation isn't the point. The point is that I never think about bookings anymore. The cognitive load is gone. My brain isn't holding "did I follow up with that prospect" in working memory. It's free. And that freed-up attention is worth more than the five hours.

The Memory: Notion AI

I've been a Notion user for years. But Notion AI turned it from a filing cabinet into an institutional brain.

Every coaching session, every client insight, every framework I develop, every lesson from a failed project — it all lives in Notion. And now I can ask it questions. "What patterns have I seen in founders who plateau at $1.5M?" "What did I learn from the last three product launches?" "Which clients responded best to the identity shift framework?"

Last month I was preparing a keynote on decision architecture and I asked Notion AI to surface every client insight I'd logged about decision fatigue in the past twelve months. It pulled forty-seven entries. Organized by theme. In eight seconds.

That's not a productivity trick. That's compound knowledge. Every observation I've ever made is now searchable, connectable, and buildable. My insights don't decay anymore. They accumulate.

The Eyes: Analytics Dashboards

I have a Looker Studio dashboard that pulls data from my site analytics, email platform, booking system, and client database into one view. Every morning I get a summary: which blog posts are driving traffic, which emails have the highest reply rates, how many diagnostic calls are booked this week, and — this is the one I watch closest — what's the conversion rate from first call to coaching engagement.

Before this dashboard existed, I had a feeling about what was working. Now I have data. And the difference between feeling and data is the difference between guessing and deciding.

One thing the dashboard told me in January: my post on the three stages of founder ceiling was driving 3x more diagnostic calls than any other piece of content. So I built a workshop around it. That workshop is now my highest-converting top-of-funnel asset. Without the data, I'd have kept writing about whatever felt interesting. With the data, I write about what actually moves the business.

The Clock: Scheduling Automation

Sounds boring. It isn't.

Cal.com handles all my scheduling — diagnostic calls, coaching sessions, speaking inquiries. But the real value is what happens around the calendar. Automated reminders 24 hours before. Automated prep materials sent to clients. Automated post-session follow-ups with action items I dictate into a voice memo that gets transcribed and formatted.

I used to lose two clients a year to scheduling friction — missed sessions, rescheduling chaos, forgotten follow-ups. Haven't lost one since I automated the whole lifecycle.

The Voice: Client Communication Automation

My email sequences are semi-automated. New prospects get a nurture sequence that's 80% pre-written and 20% personalized by me based on their diagnostic call. Existing clients get weekly check-in prompts, session recaps, and resource recommendations — all triggered automatically based on where they are in the 90-day protocol.

Here's the thing that took me a long time to accept: most of what I was writing in client emails was the same. Not identical. But structurally the same. The same encouragement patterns. The same framework references. The same next-step prompts. Once I admitted that, automating 80% of it was obvious. And the 20% I still write by hand? It's the part that actually matters. The personal observation. The specific callout. The thing only I can say because I was on the call and I heard what they didn't say out loud.

The Builder: AI-Assisted Development

I've shipped over 600 projects. The dev studio is where AI has changed things the most — and I'm going deep on that in my next post on vibe coding. But the short version: what used to require a team of four engineers for a client prototype now takes one senior dev working with AI code assistants. The speed difference isn't 2x. It's 10x. And the quality is comparable because the human still makes every architectural decision. The AI just handles the implementation speed.

This is the part that scares people. It shouldn't. The engineers who are good — the ones with judgment, taste, and systems thinking — are more valuable than ever. The ones who were mostly doing routine implementation? That work is gone. And it's not coming back. I wrote about what this means for organizational structure in The Great Flattening.

The Identity Layer Nobody Talks About

Here's where this gets real.

Your AI stack is a confession. It tells on you. Show me your tools and I'll tell you who you think you are as a founder.

If your AI stack is full of tools that help you do more tasks — more emails, more meetings, more tickets, more output — you still see yourself as an Operator. You're the person who gets things done. Your value is throughput. And you've hired AI to make you a faster Operator.

If your AI stack is full of tools that help you think better, decide faster, and remove yourself from operational loops — you've started becoming an Architect. You're the person who designs the system. Your value is direction. And you've hired AI to extend your strategic capacity.

Same tools. Entirely different configuration. Entirely different results.

Marcus — the founder I mentioned earlier — had his breakthrough when he realized his seventeen AI subscriptions were all serving the same story: "I'm the person who handles everything." Once he let go of that identity, the right stack became obvious. Six tools. All pointed at strategic amplification. None pointed at doing more stuff.

What "Team" Means Now

I want to be honest about something. My operation is leaner than it's ever been. And that's complicated.

Two years ago, I would've hired a full-time operations manager, a content writer, a social media coordinator, a VA, and a data analyst. Five roles. Today I have a senior strategist who does deep work I can't do alone, and a part-time editor who brings voice and craft to everything we publish. Two people.

But it's not that AI replaced three jobs. It's that the three jobs don't exist anymore. The work itself changed shape. What used to be "manage the content calendar" is now "approve the content calendar that was auto-generated from performance data." What used to be "compile client metrics" is now "review the client metrics dashboard and flag what matters." The coordination layer is automated. The judgment layer is human.

The people on my team do harder, more interesting, better-paid work than those five roles would have done. And they have more autonomy because the operational noise is handled.

That's what the right AI stack does. It doesn't replace your team. It changes what "team" means.

The Honest Caveats

Three things I've learned the hard way.

First: building this stack took me eight months of iteration. Not eight hours on a Saturday. I tried tools that didn't work. I built automations that broke. I had a Make scenario that accidentally sent the same email to a client forty-three times. (She was very gracious about it.) This stuff compounds, but it compounds slowly. Don't expect overnight transformation.

Second: you need enough self-awareness to know what you're bad at. I'm bad at follow-up. I know this. It's my weakest operational skill. So that's the first thing I automated. If you automate what you're already good at, you save time. If you automate what you're bad at, you fix a structural flaw. Big difference.

Third: AI is only as good as the system it sits inside. If your business model is broken, AI will help you execute a broken model faster. If your positioning is unclear, AI will generate beautiful content that says nothing. Fix the strategy first. Then build the stack.

Your Move

If you're reading this and thinking "I need to rebuild my stack" — slow down. Don't go buy seven tools tomorrow. Start with one question: what work am I doing right now that a machine could do at 80% of my quality?

For most founders, the answer is: first drafts, scheduling, data synthesis, follow-ups, and monitoring. That's the starting five. Automate those, and you'll get back 15-20 hours a week. But more importantly, you'll get back the attention those tasks were consuming even when you weren't doing them.

And then ask the harder question: am I building a stack that makes me a faster Operator, or am I building a stack that makes me an Architect?

The answer to that question will determine whether AI becomes the best thing that ever happened to your business — or just another way to do the wrong work at impressive speed.