We’re Building an App with AI Tools in Just Two Weeks—Here’s Why

We’re putting AI to the test. In public. In real time. No safety net.

Introducing the Farmstand App Experiment—a rapid-fire, two-week sprint where our team will design, build, and (hopefully) launch a fully functioning mobile marketplace app using AI-first tools like Vibe Design, Vibe Code, Figma Make, and Cloud Code.

The app idea is simple: connect small-scale farmers with local shoppers to reduce food waste and increase access to fresh, farm-grown goods. Straightforward enough. But the real story? That’s happening behind the scenes.

Why we’re doing this

This isn’t a thought experiment or a polished case study with hindsight wisdom. It’s messy, real, and unfolding live. Because we’re chasing two big questions:

1. Can generative AI tools actually deliver production-ready products? We’re not just kicking the tires—we’re driving this thing off the lot. The tools we’re using are some of the most hyped in the AI space, and we want to know if they’re all show or if they can actually ship. This isn’t a sandbox demo. This is a real sprint, with real deadlines, real users, and real consequences.

2. Can we go from zero to launch in just two weeks? Deadlines are where good ideas go to grow up—or crash and burn. We’re aiming to prove (or disprove) whether AI can meaningfully accelerate the product lifecycle. If we pull it off, it’ll be a small win for indie farmers and food shoppers. If we don’t? We’ll break down where the tools fell short and what we learned the hard way.

This is a live build

Most teams don’t share the ugly middle. We are. As often as we can, our crew will post updates—what’s working, what’s breaking, and what we didn’t see coming. You’ll get a behind-the-scenes look at what it’s really like to build a digital product with AI in 2025: the speed, the glitches, the hallucinated screens, and the moments where it all just clicks.

This isn’t a marketing stunt. It’s a working experiment. A proof-of-concept. A stress test on today’s AI tools.

Whether we launch something useful or hit a wall, we’ll share everything we learn.

Who this is for

This sprint is for:

• Builders curious about AI tools beyond the hype
• Agencies and startups looking to speed up product delivery
• Clients and collaborators who’ve asked: “Can we just use AI for that?”
  

We’ve heard the noise. Now we’re turning down the volume and tuning into the reality.

Follow along on LinkedIn. It’s going to be fast, scrappy, and hopefully insightful.