If You Can Dream It, AI Can Build It ✨

Thought Essay Business Personal Agentic AI ChatGPT Gemini Goose

We’ve all had moments with AI that made our jaw drop.

For me, it’s been a series of those moments across business and personal projects with each one raising the ceiling on what I thought was possible.

🧱 AI Building Blocks

  • 🏃‍♂️ 2022 – AI Outcoaches My Human Coach
    ChatGPT had just launched and I started using it for insights on running performance while training for the Boston marathon. I quickly realized it was giving me higher quality, more personalized coaching than the human coach I’d hired.

  • 🧑‍💻 2023 – AI Outcodes Me
    At work, ChatGPT started writing SQL and Python faster (and often better) than I could. My productivity skyrocketed, and my curiosity followed.

  • 🗂️ 2024 – AI Automates All The Things
    I started building personal GPTs using Gemini and ChatGPT to optimize my investment portfolio, coach me in fitness goals, automate content retrieval, and mentor me as a personalized AI life coach trained on my favorite quotes and books.

🚀 From Chats to Agents

  • 🛫 January 2025 – From Assisting to Doing
    Block launched an open-source agent called Goose and made the leap from AI assisting our work to AI doing the work. I learned to build AI agents using Model Context Protocol (MCP).

  • 🏆 March 2025 – Hack Week Breakthrough
    I led a team of six engineers to build a functional AI agent prototype that answered natural language questions via a Snowflake MCP to automate the analytics pipeline. Out of 300+ projects, ours was featured company-wide and voted most impactful project for data science.

  • 🚀 May 2025 – From Prototype to Product
    That hack week project evolved into a three-month sprint to build and launch an internal, AI-powered self-serve analytics engine. It gave me a front-row seat to the real challenges and opportunities of scaling AI for BI.

  • ✨ June 2025 – Vibe-Coding Takes Over
    During AI Week, CEO Jack Dorsey challenged employees to vibe-code a personal project. So I repurposed the same agentic workflows from work to my personal computer and spun up my dream Music Insights app in roughly three minutes.

That’s around the time my obsession fully kicked into gear.

Why I Built This Site

I wanted a place to think out loud, share what I’m learning, and experiment in the open.

So I’m carving out a few nights a week, after the kids are asleep, to explore, build, and write. The projects I’ll share here fall into three buckets:

1. AI For Business Intelligence

Reflections on building AI data products inside a company that’s actively shipping forward-thinking tech. I don’t speak for the company and the views and opinions shared on this website are fully my own. But Block believes in open source and shared learning, and I hope to contribute to that spirit with insights, frameworks, and lessons from the front lines.

2. Personal AI Project Demos

I’ve been using AI to build actual tools for my own life and I’ll demo these as I go not only to show what’s possible, but to show how you can build them too, even if you’re not an engineer. I’ve already got a pipeline of 10 personal projects I’ll be vibe-coding in the coming months, you can preview them here.

3. Frameworks & Thought Essays

The strategy for how we as individuals, businesses, and societies adopt and adapt to these new creative systems fascinates me. I’ll be sharing mental models and frameworks that help me stay grounded in the whirlwind of AI change. And I’ll be exploring and researching the hard questions to play my part in ensuring that we mitigate the risks that come with this powerful technology and collectively use AI as a force for good.

If You Can Dream It, AI Can Build It

When I vibe-coded my first music insights app, my knee-jerk thought was:

“This is so cool. Maybe I should try to scale and monetize it.”

Then I remembered: AI literally built it for me in three minutes. 👉 Watch it happen

That moment hit me hard. It highlighted what I think is a fundamental shift in how we’ll build and use data products going forward.

💡 The old model:
Centralized apps that store your data, give you a handful of dashboards, and expect you to adapt to them.

🔄 The new model:
AI builds highly specific, personal apps, instantly tailored to your data, your context, and your needs.

So this site isn’t just for AI hobbyists. Or data engineers. Or business leaders.

It’s for anyone with curiosity and an idea.

Because we’re entering an era where:

  • ✅ You don’t need to be an engineer to build with AI
  • ✅ You don’t need a team to launch something meaningful
  • ✅ You just need the right prompt, the right context, and a little bit of vision

Most people experience AI as a chat window. That’s still helpful. But it’s passive.

Agentic AI is different. It’s what happens when you give AI the keys by connecting it to your tools, data, and systems so it can take action on your behalf.

I’ll be exploring what that looks like, both inside a billion-dollar company and from my couch at 2 a.m. when inspiration strikes.

And I’ll do my best to document the workflows along the way so you can remix them for your own ideas.

✨ Because if you can dream it, AI can build it. ✨

- Zach

📬 Follow me on LinkedIn to get project progress updates in your feed.