GPT-3, Text AI

Interview with Stephen Campbell, GPT-3 App Builder

I recently interviewed Stephen Campbell, Creator of is a GPT-3 Powered App that lets you generate various kinds of “Copy” with just a simple prompt.


What’s unique about this app is that it’s built using a No Code tool called Bubble.


Hi Stephen. Let’s start with some background. What is your backstory?

I am Stephen Campbell, a regular Jamaican guy who has a knack for solving technical problems of any nature. I think I have always loved science and technical things from a very young age.

I used to pull my toys apart as a child with my brother and use the motors, lights, and gears to build what we called “robots” back then.

My Dad encouraged me to become a Civil Engineer because that was needed in Jamaica and he was a Property Manager so he knew it would be a good path.

I eventually became a Chemical & Process Engineer and worked in the bauxite industry for a little over three years (2015-2019) until I found a passion for machine learning and the world of internet entrepreneurship; making a living online. I eventually left that job and started on a journey to figure out how it was possible to go about building an online business.

Figuring out a path after doing WordPress site development, domain name investing and freelance independent contractor for a domain name registrar, I stumbled upon the no-code movement in August 2020 on Twitter after seeing a tweet about a “no-code” related domain name selling for $3500.

I immediately became interested in no-code and started to pursue that heavily.

What have you built with OpenAI and what does it do?

I have built quite a few demo applications with OpenAI’s GPT-3 but what has gotten the most traction is

It is a simple application that allows you to generate various Copy with just a prompt. Its use ranges from summarizing long text, answering questions, completing thoughts, and writing articles. The tool is powered completed by OpenAI’s GPT-3 algorithm.

How did you come up with the idea for your product(s)?

I came up with the idea after brainstorming for a few days on what no-code product to build. I always hated building stuff for building sake or just for practice. I always want to build something that will add value when I launch it. Even if the value is just to collect data on how people interact with what I have built.

So one day after brainstorming, I came up with the idea of somehow merging what I had a great passion for; Artificial Intelligence (Machine Learning) and no-code; ‘_No-Code AI_’ then. I was learning bubble at the time and it so happened that I had also recently applied for OpenAI beta access and was successful. So I thought I would try to figure out how to merge the two.

I shared some of the notes from my journal in tweets on the process of coming up with the idea for the product.

Why did you decide to build with OpenAI API? Have you had other exposure to AI tech in the past?

I decided to build with OpenAI’s API only because I got access to it. I never knew what it was really about outside of seeing the buzz that this might be the most powerful artificial intelligence algorithm on the planet and they were giving access to regular people.

To be honest, I am not sure why I thought I would get into the beta, because all I had was an idea and not much technical skill. I decided to apply just the same to see what would happen.

Here’s how I applied:

I had no exposure to AI tech in the past outside of trying to learn python and fooling around with word2vec, an NLP model that I found out about using Google Colaboratory (Tensorflow).

How is your product doing today? is stable right now. After launching it on Product Hunt, there was a huge spike in sign-ups, subscriptions, and usage but now it has more or less stabilized and is getting regular weekly sign-ups and subscriptions.

The live stats can be found here:

The cumulative stats to date since launch are as follows:

  • Sign-Ups: 3382
  • Paid Users: 159
  • Number of Characters Generated: ~ 8.2 Million
  • Revenue Generated: $1593.42

What are your long-term plans for the product?

I hope to keep growing this product and have it continue to pay for its own maintenance and improvement. I hope to have this as a passive income generator for a good long while or if someone offers a price that is in my range and has a higher purpose for this tool then I would consider selling.

Are you worried about Platform risk? What are you doing to mitigate that?

There is some platform risk involved with using the no-code stack I use to power the app;,,,

Most of these platforms already have intrinsically mitigating frameworks based on what I have experienced and read in their corresponding documentation. My responsibility is to ensure that I tie these together in such a way that the application is overall safe for people to use.

**What is your advice for founders looking to build products with OpenAI API? **

The advice I would give is to use your creativity to come up with product ideas. I think OpenAI has provided clear and easy-to-use documentation as well as examples of what is possible with GPT-3. The thing for founders to do is to activate their own creativity and try to come up with products that add value.

The products should improve upon what is already out there or solve a problem that no one is going after solving based on what users ask for.

Another thing is to iterate on various products and see how users interact with the product and draw new insight, ideas and find out what is necessary to succeed in the market that you are attempting to reach.

Is building an AI-enabled product any different than building without AI? What are the pitfalls to avoid while building with OpenAI?

Building an AI-related product with no-code tools is no different for me. It’s just engineering API calls with a suitable workflow.

Disclaimer: I have no experience building anything AI-related outside of the no-code space, therefore my grasp of it is very limited.

Pitfalls with OpenAI that I have experienced and also have heard lots of views from other makers and founders is the overestimation of what it can do. It is certainly powerful, but it is not absolutely intelligent. In other words; intelligence is still artificial. One should understand that the software is only AI and it is therefore limited. What you can do with it is constrained to what OpenAI has determined it can be used for. Within those pre-determined use cases, one has to use their imagination to extract the most value from it. GPT-3 is a natural language processing model, therefore it is limited to text-related prompts. It isn’t all-knowing. I think that this thought process provides a healthy balance when thinking about ideas to build with OpenAI.

Why did you decide to build your products with a No Code tool-like bubble?

I didn’t have much cash when I started with no-code and I applied for Bubble’s immerse program. They eventually gave me two months free on the Personal plan and I decided to utilize it to the full extent. I then fell in love with everything that it was about and decided to continue using it.

What kind of Tech Stack you would recommend for building AI-enabled products?


What kind of learning resources (books, podcasts, etc) you would recommend to the founders looking to build AI-enabled products?

  1. Fundamentals of Deep Learning: Designing Next-Generation Machine Intelligence Algorithms

Where can we go to learn more about you and your product?

AI Chat