No Coding Required as AIT Students Build Apps at Vibe-athon

Imagine being able to build your own app without any technical background or programming experience. That is what 19 teams of AIT students discovered at AITSPIN’s Vibe-athon on 12 September, where they used AI-powered app builders to turn prompts into working business applications.

By the end of the event, Haiqa Jalil and Momina Bibi won on the Lovable platform with an app they called “Corolla Wings Invoicing”, while Bishu Khanal and Sameer Chapagai won on Bolt with “Yeti Orders”.

Conceived jointly by AITSPIN, the SmartDATA Lab and the AIT Student Union, the Vibe-athon was developed under the guidance of Dr. Chutiporn Anutariya and organised by her master’s student, Pratibha Hamal, as part of Hamal’s thesis. Dr. Chutiporn also delivered the opening remarks. The hands-on event challenged participants to turn a business problem into a functioning application.

“AI is making it possible for people without a technical background to build things that previously required programming skills. The challenge now is to learn how to use these tools thoughtfully, critically and responsibly,” said Dr. Chutiporn Anutariya.

Participants could use two AI-powered app-building platforms, Lovable and Bolt.new. Instead of writing code line by line, they described what they wanted in plain language. They then reviewed the results, tested their applications and continued refining them through additional prompts.

The central task was to build an invoice management application for a fictional small business called Corolla Wings. The application had to accommodate different users, including the business owner, staff, an accountant and customers, each with different levels of access and authority. The owner, for example, was the only user authorised to give final approval to an invoice.

With AITSPIN’s Noppharat Nuamkhein providing technical support throughout the event, participants had three hours to build and publish their applications. Teams were free to prompt the AI builders as many times as needed, make changes, test their applications and improve them.

The applications were judged on completeness, user experience and design, practical features and output, and overall impression. The judging panel comprised three PhD candidates in the Data Science and AI Program: Suhel Acharya, Rakshya Lama Moktan and Parkpoom Wisedsri.

“A working application is not necessarily a secure one. An app can look polished and function correctly, but still contain security weaknesses or other problems,” Pratibha told the participants.

The Vibe-athon formed part of Pratibha’s master’s thesis on how non-technical users build applications using AI-powered tools. Her study examines how participants formulate prompts, refine their instructions and develop the applications they eventually produce. It also provides an in-depth analysis of security vulnerabilities in the resulting vibe-coded applications, and investigates methods for detecting and preventing these vulnerabilities.