
----------------------------------------
LACORTE v 1.0.0
(c)2025 LACORTE Industries
- Server 591 -
----------------------------------------
January 05, 2026
Let me tell you about the time I thought I was a real developer. It was 7 years ago. I was young, naive, and full of... well, mostly caffeine and questionable life choices.
I created an NPM package. A real one. Published it and everything. It's called alphavantage-sdk, and it's still out there, haunting the NPM registry like a digital ghost from my past.
I wanted to use the Alpha Vantage API (which provides free stock market data, because who doesn't love free APIs?). But making HTTP requests was hard, apparently. So I did what any reasonable developer would do: I wrapped it in a library and published it to NPM.
Because why use fetch or axios directly when you can add an extra layer of abstraction? That's what good developers do, right? Right?
The package is simple. Almost embarrassingly simple. It's basically a wrapper around Node.js's built-in https module. But hey, it worked. And people actually used it. Some still do, probably. God help them.
The entire SDK is... wait for it... about 50 lines of code. That's it. 50 lines. For an entire NPM package. I'm not proud, but I'm also not ashamed. It's a weird middle ground.
The code does exactly what you'd expect:
I used Node.js's built-in https module because I was too lazy to add a dependency. Which, looking back, was actually a smart move. Fewer dependencies = fewer problems. Who knew?
The code uses Promises (because async/await wasn't cool yet in 2018, or I just didn't know about it). It manually concatenates response chunks because... well, because that's how you did it back then, I guess?
The SDK provides functions to fetch stock market data:
It's basically a thin wrapper that:
That's it. No magic. No complex logic. Just HTTP requests wrapped in functions. And you know what? Sometimes that's enough.
I published it to NPM. People installed it. Some might still be using it. I'm sorry, future developers who have to maintain code that depends on this.
The package name is alphavantage-sdk. It's still there. You can install it right now with npm install alphavantage-sdk. Please don't. But you could.
It has a README. It has examples. It even has a CONTRIBUTING.md file (because I was fancy). It's licensed under MPL 2.0, because I thought that was a good idea at the time. I still think it is, actually.
Let's be honest: this code is not impressive. It's a simple HTTP wrapper. Anyone could write this in 10 minutes. I probably did write it in 10 minutes. But I published it, and that's what counts, right?
The code uses:
https module - Because dependencies are for the weakIt's not good code. But it's functional code. And sometimes, that's enough for an NPM package.
Looking at this code now, I see all the things I'd do differently:
But you know what? It worked. People used it. And that's more than I can say about some of my more "polished" projects that never saw the light of day.
This package is old. Really old. 7 years old. In JavaScript years, that's basically ancient history. The code is simple. The approach is outdated. The implementation is... well, it's an implementation.
But here's the thing: I shipped it. I published it. People used it. And that's something. Not everyone can say they've published an NPM package, even if it's just a simple wrapper.
It's not impressive. It's not complex. It's not even that good. But it's mine, and I'm weirdly proud of it. Even if I'm also slightly embarrassed by it.
If you want to see the actual code (and judge me for it), you can check it out on GitHub. Or install it from NPM if you're feeling adventurous. Just don't tell me if you do. I don't want to know.
Tags: [ portfolio ] [ nodejs ] [ npm ] [ api ] [ retro ]