Obviously, lack of inspiration for the title… especially after 1 year from my last post 🤣
Right, so what have I been up to? Not much, radio-wise. Got a new puppy, a Romanian Raven Shepherd, named Onyx. Amazing creature! Big one as well, he’s only 8 months old now and over 30Kg.
Back to the post: like any software engineer nowadays, wanted to see for myself what’s the big fuss with all this AI stuff, and decide whether I should cr*p my pants (pardon my French…) about losing my job soon or not.
And although I do see massive advancements with this, with all the tools available now, think there’s still some time left for us, “old school devs” at least. I mean they work great, but no chance of me pushing any (more complex) AI code into production without proper supervision, reviewing the code and all that.
It does help with the heavy-lifting indeed, getting out of the way repetitive tasks, finding patterns/bottlenecks faster than any human (probably) would, making you waaaay more productive.
And you have to be extremely precise with the initial requirements otherwise, if you’re more vague, will spend a looot of time correcting and re-doing stuff.
So, I got into this “vibe-coding” trend all the cool kids are doing nowadays. Using the DeepSeek V4 Flash model I started to build (initially) an FM SDR receiver that works with the input from a rtl_tcp source.
In the meantime, added NFM, AM and DAB functionality to it, as well as support for multiple accounts, all running in a docker container.
And I didn’t write A SINGLE LINE OF CODE!!!!
The project is available in GitHub if you want to check it out: https://github.com/georgeschiopu/fm-dab-radio-receiver
That’s it from me for now. Probably my next post gonna be in about a year or so…
73