What happened so far
Over the last few days I started building "in public" by posting regularly on X. If I measured success by the reach of those posts, I'd call the results unexciting. But that isn't the plan. Of course, I dream of waking up to an account flooded with likes, but that's the dream, not the plan. The plan was to go from dreaming about building something to actually delivering it.
If the measure is delivery, then I'm pretty satisfied: I'm starting to have something that kind of works. It has a ton of limitations and issues, but it's already useful to me and, soon, I'll let friends and family test it too.
In other words, that simple daily recap of my work gave me consistency and focus, helping me carve out time in a very busy life. And it feels great: I can see something I've built, and I've watched far fewer YouTube videos.
Why a masterplan now?
I know I'm late: as I write this, there are fewer than 80 days left in 2026. But that's actually perfect for a consistency experiment I can look back on at the beginning of 2027.
What I find most difficult when building alone is staying consistent. A structured environment imposes a pace and a set of objectives. Working alone is the opposite: there's no discipline, no system other than the one you build around yourself.
So this post is part of the masterplan itself: I'm writing down a reasonable amount of work that I can fit into my free time over the next few months. That's also why I'm not worried about anyone stealing my ideas. I'm writing this to hold myself to the plan and, if someone tries to build the same thing... well, ideas are cheap, execution is everything.
Social media and I
To be clear: I think platforms like YouTube, X and Reddit are great... but I've often wasted far too much time on them. It's far too easy to end up mindlessly consuming content and burning through what little free time I have.
When you're creating or building, the paradigm changes: you have to think, step back and look at what you're doing from the outside, test it and see whether it works. I'm not saying my creations are great, or even good. But the process feels great to me, and it pushes me to learn and stretch my limits.
That's why my first objective is to create more and consume less. If I have to put a number on it, I'd like to consistently post at least two YouTube videos a week (mostly shorts). Now that I have a microphone and new lights (which I still have to test), this should be possible without spending an hour setting up the stage every time.
As for X, it would be great to keep the current pace: at one or two new followers a day, I'd love to reach 150 by the end of the year.
Why am I doing this? A bit of vanity, I guess, but also to connect with like-minded people and build a community that might one day use the services I create.
Building a SaaS might not be the most original or glamorous idea, but I really enjoy building it.
InFocusBlocks and other projects
In my posts on X I've mentioned a project I've been working on: an app called InFocusBlocks. It's just a temporary name: I'll come up with a definitive one soon enough.
The idea is fairly simple: it parses your study material (lectures, PDFs, notes) and organises it so that an AI can quiz you and create personalised lecture notes. As a next step, I'll add a spaced repetition algorithm to help you review what you don't know yet.
The name comes from how I plan to use it: I'll answer the questions on an e-reader or on printed sheets, focused, in fixed blocks of time... and then send my answers back to the app to have them graded and reviewed. This is what I call "slow AI": the idea that, sometimes, the best use of AI isn't a direct, continuous interaction but an asynchronous job that leaves your mind time to think and learn. Because if you want to truly concentrate, reflect and learn... well, you need blocks of focused time.
I'm happy that it's slowly coming together. It isn't a particularly innovative or exotic project, and it doesn't feel like I'm building the thing of the future. But it's a great start: it shouldn't be too complex, and I can actually use it myself. After failing a bunch of times, I'll hopefully build the cool app that looks like it came from the future too. One thing at a time.
So far, most of the parsing and quizzing is done. Today I even tried to host it at home on a Raspberry Pi 3: unfortunately, the SD card I'm using seems to be too slow, so for now I'll serve it from my laptop. After some more testing, I hope to open a beta to friends and then to anyone on the internet who wants to try it.
During this phase I also want to add the spaced repetition features. Then the plan is to launch it as a SaaS by the end of the year, or to nuke it if no one is interested.
Still, I believe there's a niche for "slow AI": using AI not to pull us into an artificial world, but to enhance our real one. In this case, with a virtual tutor that's there for you 24/7.
Of course, I have a lot more ideas. One stream of ideas I'm trying to avoid is optimisation: I could use different models or other solutions, or run everything on my Raspberry Pi 3 to save energy. These ideas aren't bad. But the biggest challenge is getting something into people's hands and getting feedback. That's what has the greatest value: feedback, and tuning things until they're just right. Saving on the backend or making things a bit faster matters, but it isn't fundamental.
I believe feedback is one of the only real moats a SaaS has. Sure, anyone can give an LLM a prompt and get an app that does all this in a matter of hours... but is it something you'd actually want to use? Of course, if I were you, I wouldn't trust me either: I haven't launched a single SaaS. Not even one that failed.
Many of my other ideas are about new projects. Next in line is one for my children: an app to help them learn English and German. The only reason I haven't started it yet is that it would make me lose focus. Once this first SaaS is up and running, I'll move on to the next one.
This is also the first time I'm building a whole product from scratch, on my own: the app itself, but also accounts, hosting, deployment and keeping everything secure. That's why a single project feels like more than enough for now. I'd be happy to end the year with my first SaaS launched and a new one underway.
Want to beta test InFocusBlocks? DM me on X or reach out through any of the channels on the About page.
Recap: the plan until December 31st
- YouTube: post at least 2 videos a week (mostly shorts).
- X: keep posting daily and reach 150 followers.
- InFocusBlocks: open a beta to friends, then to the public.
- InFocusBlocks: add spaced repetition.
- InFocusBlocks: launch it as a SaaS, or nuke it if no one is interested.
- Next project: start the English/German learning app for my kids.
I'll check back in early 2027 to see how it went.
Let's strive together.
GioStrives