Test-n-Tell - tiny apps, Big ideas

  The beginning of development went surprisingly smoothly. My game concept was simple and well understood, so Codex had no problem getting the project started. It created the basic structure in Godot , set up a playable scene and implemented the core rules. Very quickly, I had something resembling the main gameplay screen where I could move balls around the board and see whether the basic mechanics actually worked.

In this series of posts , I am sharing my experience creating a game from absolute zero as a non-developer and non-designer. In the introductory post , I covered why I picked a game for this experiment. Today, we'll go through the first steps of the process: selecting and setting up the key tools.

My blog's stats are clear: people are interested in practical experiences. Especially when it comes to AI and vibe coding tools. Luckily, that's what I've been doing these past few months - heavily using AI to create a digital product end-to-end. This series will get into all the fancy and stinky details of my process, challenges I've encountered along the way and, of course, the final result accompanied by the cost breakdown.

  Like it or not, lots of people are trying vibe coding. App Store and Google Play are flooded with new apps and games. Product Hunt launches, and domain sales are rising. But is any of it good? And does it matter?

  Nir Eyal’s Hooked made a huge impression on me. It shaped a large part of how I think about product creation. Not because I wanted to “hook” users at any cost, but because the book made one thing very clear: if you understand how to influence behaviour, you also become responsible for how you use that power.

  I have been using Codex more and more recently. Not in a “look, AI wrote me a todo app” kind of way. More like: I need to ship my actual product, fix real bugs, improve real tests, and not spend my evenings arguing with a robot about why it deleted half of my codebase.

  An AI agent is like a super obedient engineer. It does whatever you ask. It rarely challenges you. It rarely pushes back. It rarely says: “This is a bad idea.” And if it goes off script, well, that’s probably because your instructions were too vague. Finally, executives got what they always dreamed about: a builder who simply builds.

The job of a product manager is already changing. We use AI more and more in our daily work, and there is no sign of this slowing down. Writing, research, analysis, prototyping, summarising calls, checking requirements, preparing strategy docs, testing product ideas - all of it can already be supported by AI.

  When I started vibe coding, I had a budget in mind. I knew I needed to learn a bunch of new AI tools, and that would cost something: subscriptions, hosting, third-party tools, APIs. And it did cost me a bit… then a bit more… and finally a lot more. So here’s me sharing a few things I learned (and paid for) so you don’t have to.

  I've been doing a lot of vibe coding lately, and one tech stack emerged for me as the most practical. Spoiler alert 1: it's Firebase + Codex + Google Cloud. Spoiler 2: tomorrow it could change.

“Vibe coding”, building software by describing intent and letting AI generate, refactor, and orchestrate code, is quickly evolving into something more powerful: agentic coding. Systems that don’t just autocomplete code, but reason across files, plan changes, run tests, deploy, monitor, and iterate. If this trajectory continues, we are not just accelerating engineering. We are reshaping the entire product value chain.

  As AI moves from experimentation to production, many organisations discover that “adding AI” is not a feature - it is a fundamentally different product problem. This realisation has led to the rise of the AI Product Manager (AI PM).

  Everyone who has ever tried to create a product has their embarrassing failure story. Here's one of mine.

  If I'd try Jules first - I'd have a totally different impression of the state of coding agents today. A much worse impression. Here's why.

  I have built products in many different industries: HRtech, marketing, last-mile.. But one industry, where I really wanted to work, always eluded me - that's Education tech. The stars were never rightly aligned for me to work in EduTech until I decided to build my own EduTech app.

  Ok, what's now? On Saturday, I celebrated my 2000th day streak on Duolingo. How long is that? Almost five and a half years using this app every day! But that Saturday was the last day I used Duolingo.

  People come to product management from different fields, engineering being just one of them. This created a divide between technical and non-technical PMs. With the rise of AI coding tools, this divide is going away fast, as most PMs can now become technical PMs.

  In this series, I've tested and reviewed multiple tools in the AI agents category. Some of them are useful, others less so, but one thing is certain: this category has a huge potential to change our whole industry and have major effects on people's roles.

  Writing is an essential PM activity - depending on the type of PM you are, you either write a lot or a lot a lot. Many PMs these days employ gen AI tools to help them write more or better, but they do not always get the outcomes they hoped for.

  Have you ever wished for AI to only answer based on the data you give it? Or maybe you want to turn some of your sources into a real-sounding podcast? This and much more is possible with NotebookLM.