Case study

How I Built Daily Devout Readings

A friend asked me how I built this. I told him I used Cursor AI. Here's how I did it.

SaaS AI-built Devotional

Overview

Daily Devout Readings does one simple thing: it gives you one short spiritual reading per day, drawn from classic devotional books — the kind St. Francis de Sales recommended. Augustine, Bonaventure, and others. You don't have to pick a book or remember where you left off. Each day you get a single, manageable snippet. You can read it yourself, or have it emailed to friends so they get the same daily habit.

The idea

The idea came from a real problem: those classic devout books are long and easy to put off. St. Francis de Sales said to keep "some good devout book at hand," but it's hard to build a habit of "a little every day" when the book just sits on the shelf.

This project tries to solve that by serving one snippet per day, from a rotating set of those classic works, so the habit is small and steady instead of overwhelming.

How Cursor AI helped build it

I didn't build this from scratch alone. I used Cursor, an AI-powered code editor. Here's how that worked in plain terms.

1

Describing what I wanted

Instead of writing every line of code myself, I could describe the goal in normal language. For example: "I want a program that picks one paragraph per day from a list of books, and the same day should always show the same paragraph so people can share 'today's reading.'" Cursor's AI understood that and helped generate the logic to do it.

2

A pair programmer that writes code

Think of Cursor as a pair programmer that never gets tired. I'd say things like "add a way to send today's reading by email" or "make it so we can list all the available books." The AI would propose code and explanations, and I could accept, tweak, or ask for changes. That made it much faster to go from idea to working tool.

3

Fixing and improving as we went

When something didn't work or I wanted a different behavior, I could describe the problem: "The email isn't sending," or "I want the daily choice to be based on the date so it's the same for everyone." The AI would suggest fixes or new code, and we'd iterate until it did what I wanted.

4

Adding features without starting over

Over time I added more features: downloading and caching the texts, configuring which books are in the pool, sending the daily snippet by email, and later optional sharing to X (Twitter). Because I could describe each new feature in plain language, we could layer them on without throwing away what was already built.

Why it matters

You don't have to be a programmer to get the point: Cursor let me focus on the idea and the behavior I wanted, while the AI handled a lot of the detailed code. I still made the decisions — which books, how the daily pick works, how email is set up — but I didn't have to write every line from scratch.

That's how this project went from "I want one devout reading per day" to a real tool I can run and share.