The $1.5M App With No UI: How One Developer Built a Tech Empire in a Hidden Market

Most developers today are fighting over the exact same trending software ideas, completely unaware of the massive opportunities hiding in plain sight. One software developer, Jordan, solo-bootstrapped an app that has generated over $1.5 million in total revenue—and it doesn't even have a user interface or a mobile app.

Here is how he found a customer base nobody was building for, validated his idea, and turned it into a highly profitable SaaS business.

The Accidental Discovery

Jordan’s journey didn’t start with a traditional startup pivot. After working as a researcher for a Fortune 100 company and writing software for self-driving cars and moon rovers, he transitioned to freelance engineering in San Francisco.

The "aha" moment came unexpectedly. Jordan was finishing up a mobile app for a client when he jumped on a Zoom call, only to be greeted by the client's girlfriend—who informed him that his client had just been sent to prison. Jordan kept in touch with him via paper letters and quickly learned a hard truth: existing tech services for incarcerated individuals were a massive scam, offering low-quality functionality at exorbitant prices.

Enter Parakeet Chat

Seeing a glaring problem, Jordan decided to build a better solution. He created Parakeet Chat, an AI learning and communication tool specifically designed for incarcerated people.

Because of the strict technological limitations inside prisons, the app operates entirely without a screen-based UI. Instead, it functions as a chatbot integrated directly into the internal prison email system. Users email a specific address, and the bot processes the request, allowing them to:

• Interact with ChatGPT to study case law and research their legal rights.

• Look up sports statistics.

• Ask entrepreneurial questions to prepare for life after release.

• Send messages to their families on the outside.

In a clever twist on the standard SaaS model, the "users" (the incarcerated individuals) are not the "customers." The families on the outside pay the $15 to $20 monthly subscription fee to keep the service running.

The Scale of a Niche Market

By solving a highly specific problem for a hyper-niche community, Parakeet Chat exploded. It currently makes over $300,000 a year in revenue.

The impact is staggering: around 30,000 people have used the platform, which accounts for roughly 20% of the entire federal prison population in the United States. To date, the system has sent out 9 million messages and facilitated nearly 100,000 family connections that otherwise wouldn't exist.

The Tech Stack (And Why You Shouldn't Care)

Jordan built the platform using TypeScript, React, Postgres, Redis, Auth0, Prisma, Zod, and Docker. But when asked about what language aspiring founders should use, his answer is incredibly blunt: "It doesn't really matter... the tech stack is not that important as sheer speed".

In fact, Jordan notes that AI is essentially the new tech stack. He now uses AI systems to write almost all of his code, admitting that he hadn't even opened his code editor in three to six months.

Growth, Validation, and the "Scientist" Mindset

Because prison is a closed ecosystem, traditional marketing strategies like landing pages and email lists were useless. To validate the idea, building the MVP and validating it had to happen at the exact same time. Jordan got his prototype directly in front of his contacts inside the prison, asked for feedback, and let word-of-mouth take over.

podcast

He also built an internal referral system offering free credits to users who brought in paying customers. Within just one month, he had 200 paying users and achieved profitability.

For developers sitting on a backlog of ideas, Jordan offers a vital piece of advice: stop being too emotionally invested in your ideas, and accept that you must validate them.

"You need to start f***ing up right now," Jordan advises. Instead of spending 12 months perfecting an app nobody wants, he encourages builders to act like scientists. Build an idea—even a stupid one—let the world hate it, let it fail in a controlled way, and reiterate based on the data you get back. As he puts it, you will learn more from that mess up than you will from reading 20 books.