One of the most interesting things happening right now is how quickly the barrier to software development is disappearing. Tools that would have required a developer, a big budget, and weeks of work not very long ago can now be built in an afternoon. We have talked about this a lot in prior tiny AI labs.
This week's challenge is to experience that shift firsthand building our own phone app in just a couple of hours.
For this challenge, instead of your usual AI tool I encourage you to use an AI app builder such as Buildy, Lovable, Bolt, Replit, or another tool that can create a live application. One of the interesting things about these tools is that your project can live beyond the chat window. Instead of generating code that sits in a conversation, you end up with a working application that has its own URL. You can save it to your phone, open it whenever you need it, share it with other people, and continue improving it over time.
Before we start: if you are getting this email forwarded to you and are not yet part of the lab - you can join the tiny AI lab here.
The Challenge:
This week, we're going to build an app that has it’s own URL.
Your goal is to build a simple app that solves a real problem for you and earns a permanent spot on your phone. The app does not need to be complicated. In fact, the simpler the better. The best projects are often the ones that remove a small annoyance, help you stay organized, or make a task easier to complete.
Here are a few ideas to get you started:
A reading tracker where you can log books, save notes, and capture key ideas so you can remember what you thought about them later.
A meal planner and grocery tracker that helps you decide what to eat during the week and automatically generates a shopping list.
A decision journal where you can record important decisions, your reasoning at the time, and revisit the outcomes later to improve your judgment.
A work wins tracker where you can record accomplishments throughout the year and quickly pull them together for performance reviews, promotion packages, or job applications.
You can build one of these ideas or come up with your own.
The prompt to get you started
Choose an app builder you want to try and use the following to get you started:
"I want to build a mobile app that helps me [describe the problem].
The app should have a clean interface, save data between sessions, and be simple enough that I would actually use it regularly. Before building anything, ask me questions about how I want it to work.
Once the AI starts building, continue the conversation. Ask it to improve the workflow, simplify the interface, or suggest additional features. Try to keep it simple with one or two features to start.
In my experience, the quality of what gets built depends much more on the conversation than on the initial prompt.
A Quick Security Note
As these tools become more powerful, it is worth being thoughtful about what you connect to them.
For this challenge:
Avoid using confidential work information.
Avoid connecting production databases.
Avoid uploading sensitive customer data or private information - even in conversations not going into code.
Be intentional about what personal information you store in your application.
Review what permissions you are granting when connecting third-party tools and accounts.
I would also encourage you to do your own research on how the tool you choose uses your data and what type of security they have in place to prevent unsafe data use.
Too easy?
Most AI-generated apps start with a basic interface and a few core features. Once you have a working version, push it a little further. Try asking the tool to generate summaries and insights from your data or Add AI-powered recommendations based on the information you collect over time.
"Review the data collected in this app and suggest ways AI could generate useful insights, identify patterns, summarize information, or make recommendations that would help me make better decisions."
Why this matters?
App development is getting easier and easier.
For most of software history, there was a huge gap between having an idea and having a working product. Today, that gap is shrinking quickly. That doesn’t mean we are all about being software developers - I have actually tried and failed to replicate some of my favorite apps.
What it does mean is that more people can start creating tools that fit the way they work instead of adapting themselves to existing software. We are moving toward a world where software becomes much more personal. Historically, most products needed to serve thousands or millions of users to justify the cost of building them. As a result, they were designed around common use cases and average users.
Increasingly, it is becoming possible to create tools that reflect your own workflows, preferences, and quirks. A meal planner can be designed around how your family actually eats. A reading tracker can capture exactly the information you want to remember.
For a long time, we have adapted ourselves to software. If an app didn't quite work the way we wanted, we either changed our process or learned to live without the feature we wished existed.
We are moving toward a world where it is increasingly possible to create software that adapts to us instead. That doesn't mean AI can recreate every application. Complex products with years of development behind them are still difficult to replicate, and I've run into plenty of limitations when trying. But the capabilities are improving quickly.
I think it's worth spending some time experimenting with these tools to understand what is possible.
That’s it for this week.
Hit reply and tell me what you thought of the challenge this week.
I read every response, and your experience helps me shape these challenges. If you get stuck or have questions, reach me at [email protected]