When I started building DevRecall, I didn't have a defined vision in mind.
There was no plan to launch a product, no roadmap, nothing like that.
I just had a pretty simple problem: I felt like I was getting rusty and available tools felt not really built for what I needed (we'll come back to this later).
At that time I was going through interviews again, and it became obvious - if I don’t practice regularly, I slow down. Not dramatically, but enough to notice it. And in interviews, that matters.
So I thought: alright, I should probably build something.
And definitely it should be not another TODO-list or other "classic" tool for coding practice - I wanted to make something valuable, just didn't know what it should be.
But very quickly I ran into a second problem, and it turned out to be more important than the first one.
Every interview cycle felt the same. I kept revisiting the same topics, googling the same concepts, asking AI the same questions. I had notes, but they were scattered - some in docs, some in bookmarks, some just lost somewhere.
Nothing was structured. Nothing was easy to come back to.
That's the moment when DevRecall started to take shape - I wanted a place where all this knowledge actually connects and grows over time, instead of disappearing between tabs and tools.
So I started building DevRecall as something I would genuinely use myself.
And somewhere along the way, something shifted.
Coding stopped being the most interesting part.
Not because I stopped enjoying it - I didn't. But it became just one piece of a much bigger picture.
I found myself thinking more about things I used to barely touch:
How should this product actually be structured?
What channels to use for a marketing strategy?
Is the product in good shape from a security perspective?
What does GDPR even mean for something like this?
None of that was part of the original plan. But each of these questions pulled me deeper, and weirdly enough, made me a better engineer.
AI tools amplified that shift even more.
Working with tools like Cursor or Claude, I can move much faster than before. Prototyping something that used to take days now takes hours.
But there's a flip side to that speed.
When everything becomes easier to generate, it also becomes easier to skip understanding. To accept solutions too quickly. To move forward without really questioning things.
At some point I realized that the bottleneck is no longer "can I write this code?"
It's "do I actually understand what should be built and why?"
That's probably the biggest shift for me.
From writing code... to owning outcomes.
It's not just about implementing a feature anymore. It's about deciding if it should exist at all, how it behaves in real life, how users interact with it, and whether it solves anything meaningful.
And that's where most of the growth happened.
I'm still far from having a "perfect product". DevRecall is very much a work in progress.
But even at this stage, the side effects are already there:
I think differently about systems.
I pay more attention to trade-offs.
I'm more comfortable operating in uncertainty.
Things that are really hard to get from tutorials or courses.
If you're a developer and you've been thinking about building something of your own - I'd say just try it.
It doesn't have to be big or ambitious.
Right now, with AI tools, the barrier to entry is probably the lowest it's ever been. You can experiment quickly, validate ideas, and iterate without overcommitting.
Worst case - you learn a lot.
Best case - you build something meaningful.
I started this project just to stay sharp at coding.
Ironically, coding became the least interesting part.
And that turned out to be the most valuable outcome.
Top comments (0)