Hello Thrivers,

If you have ever used Canva to quickly create a design, Nigerian developer Akinkunmi has built an open-source alternative of his own. It is called Avnac, which is Canva spelled backwards, and it started not as an attempt to take on the design giant, but from Akin’s curiosity about how graphics work on the web.

But Avnac is only one part of his story. Akin started building websites on his mother’s phone at 13 and, before 25, has built more than 11 products across developer infrastructure, cloud deployment, design, automation, and finance.

We spoke with him about building Avnac, the many other products he has shipped, why he builds so quickly, how AI has changed his approach to software development, and what he thinks will still separate great engineers in an AI-powered future.

Here is Akinkunmi’s story, in his own words.

I started my software journey when I was 13, but at the time, I did not even know I was getting into tech or that software development could become a career. I was simply curious about how computers worked and how games were created.

There was a tool called Wapka that was a little like WordPress and allowed you to create and customise websites without needing a laptop. That was what I used to create my first couple of websites. Eventually, I started learning how to code through a website called HTML.net.

Even then, I wasn't thinking about making money from software. I didn't know that was something I could do professionally. I was just interested in understanding how these things worked and building them myself. 

I built an alternative to Canva because I was curious, not because I wanted to compete with it. I wanted to build something that would force me to work with those technologies, so I built Avnac.

That is actually how I learn.

Some of my products come from problems I personally have, but others exist simply because I am curious about something. When I want to learn, I don't like just reading or cramming information. I want to get hands-on and build something with it.

That was Avnac.

Beyond Avnac, I have built products across developer tools, finance, automation, and infrastructure.

One of my favourites right now is OutRay. Normally, when developers build a website or application and want to put it online, they deploy it through platforms like Vercel, Netlify, Cloudflare, or AWS. But sometimes you just want to quickly show somebody what you are working on without setting up an entire deployment process.

OutRay turns your computer into the server. You can make changes locally and have them available online immediately.

Then there is ByteShip, which provides file upload and storage infrastructure. If you are a small startup building an application that needs users to upload images, videos, or other files, you may not want to build the entire media infrastructure yourself.

I also built Money Page, which lets you upload bank statements and generates a report showing things like where your money goes, who you send the most money to, and how much you have earned over time.

There is Envii, which solves a developer problem around securely moving environment variables between computers, and Fettino, one of my earliest products, which helps developers and designers work with colours, gradients, palettes, and different colour formats. Fettino was actually the first product I made my first $1,000-plus from.

I built Message Me, an anonymous messaging product that got around 30,000 users before it shut down, and Plaything, a real-time tic-tac-toe game where you can play with friends, strangers, or AI. That got around 10,000 users.

There is also Rockseed, which works like Discord or Slack for online communities; Aeroplane, which helps developers deploy applications on their own servers without having to build the deployment infrastructure themselves; Fireflow, an AI automation platform; Formdrop, which lets developers add forms to websites without building the backend; and Better Secrets, which provides a better interface for managing secrets used in GitHub Actions.

One of my more recent products is Revindex, or Revenue Index. It lets a business connect supported payment providers so revenue claims can be verified against actual payment data.

I've built other things but never launched them, so I don't really count those.

People sometimes assume I must have some special workflow or templates that allow me to build quickly.

I don't.

Every project starts from a blank folder. I initialise whatever framework I want to use and start building from there.

My “secret” is basically sitting in front of my laptop for many, many hours. I probably spend more time in my chair than I do in my bed.

People spend a lot of time asking what workflow I use, but that time could be spent building. You just have to sit down, lock in, and continue coding.

That is really it. 

My main goal has always been to get an idea out of my head and turn it into a working product as quickly as possible.

That has not changed with AI.

Before AI, I could still ship products in a few days because I would sit at my laptop and code for hours. What AI has changed is how quickly I can move.

There are things I can now build in less than 24 hours that might previously have taken me one or two weeks, and that would have been one or two weeks of coding for 10 to 14 hours every day.

So fundamentally, my approach is the same. The difference is speed. AI has compressed the amount of time between having an idea and having something that actually works. 

With AI, almost anybody can ask a model to build an application. But the difference is whether you actually understand what it is doing.

Can you explain the architecture behind your product? Why did you choose a particular database? Why was a feature implemented in a particular way? When something breaks, do you know how to investigate the problem?

Take observability. If you simply ask an AI to build an application, it may build what you requested. But if you don't know that you need systems to monitor what happens when that application fails, you may never ask for them.

Scaling creates the same problem.

Imagine you are building a health application that receives new data every five minutes. A normal database might work perfectly with 100, 1,000, or even 10,000 users. But as the system grows to 50,000 or 100,000 users, the architecture may need to change.

AI can help you make that change. The real question is: do you know that you need to make it?

That, for me, is one of the differences between someone who can use AI to build something and an actual software engineer.  

I don't even know what software development will look like in the next one or two years because the way we build software is changing so quickly.

Not long ago, even when developers used AI, we still spent a lot of time reviewing the code it generated and correcting mistakes. Today, there are developers, including me, who can spend much more of the development process interacting through a chat interface rather than constantly working directly inside a traditional code editor.

Maybe in another two years we won't need to see much code at all. Maybe eventually applications will be expressed more through plain English. I am exaggerating, but that is how difficult the future is to predict. 

Hopefully, by the time I am 30, OutRay will have become a proper company worth millions of dollars.

If that doesn't happen, I would like to work somewhere that builds infrastructure at a very large scale; companies like Cloudflare or Vercel interest me. Even then, I know I would still be building my own projects on the side.

I don't want to feel like a cog in a machine. If I work at a very large technology company, I would rather be somewhere I can experiment, take ownership, and clearly see the impact of what I am building.

But what I would prefer most is running a profitable startup of my own, whether that is OutRay or something I haven't thought of yet.

I have already built products used by tens of thousands of people. One of my goals is to eventually build something used by hundreds of thousands or millions.

Outside software, I want a beautiful family, to travel, enjoy life, and help people. Hopefully, I will have started a foundation by then. I am also curious about building traditional businesses that have nothing to do with software, just to see what applying my way of thinking to a completely different industry would look like.

But as an engineer, the answer is much simpler.

I just want to build software, enjoy what I'm doing, and make good money from it. 

Akin was building long before AI. AI has simply made him faster. His story shows that as software becomes easier to create, understanding how it works, scales, and solves real problems still matters. For Akin, the goal remains simple: get the idea out of his head, build it, and ship it.

Till next time,

Team Thrive

_____________________________________________________

A few lines to keep you motivated, going, and on top of the world