This...is my life. Not necessarily finding stuff on HN and pursuing them, but just projects in general.
Here's how I deal with that:
I don't start a project for a while, I think about how I'm going to implement it in pretty good detail in my head.
Then, I move on to a project I've already started, or a new idea.
After a few days/weeks/months, I'll cycle back, order parts if it's hardware, or scaffold out an application, or start writing the big picture idea of a podcast. (I've been remodeling my house, and writing a podcast for 2 years, work still continues, but I bounce between them.)
Once I have the parts/materials/scaffolding, I bounce between projects usually dedicating days to weeks on them, until I get bored or frustrated.
I've been doing this since I was a kid. I've started two businesses, written a few Android apps, I've remodeled half my house (so far, windows, drywall, and insulation are happening next week) in the past 2 years.
There are 3 projects I've been toying in my head with for a few years that I've been struggling to figure out how to start. Fully automated made to order vitamins, a backpack that has the features I want, and a "life logging" app that doesn't suck. I'd be very happy if someone else made those, I don't care about the money from them, they are things I want to exist in the world.
The made to order vitamins, costly and risky to start. The backpack I've attempted and broke the shit out of a sewing machine. The life logging app Ive started like 3 times, first on Android, then on Django, now I'm leaning towards serverless with react native...but the scaffolding for what I want is complicated enough where it's overwhelming to think about if I'm implementing it right for scaling and expandability.
Either way, the point is, I cycle through projects, making sure I leave them in a state where I know what needs to be done next. Having 2-3 outside of work projects and 2-3 work projects keeps me busy and just disorderly enough where I don't get bored and if I get frustrated, I can just switch what I'm working on.
The key is, the projects have to be different enough where it doesn't feel like you are going back to the same grind. Programming, construction, design, and podcasting are very different from one another. I'm not an expert in anything really, but I can run a software team, look at the big picture, find pitfalls, and I'm always down to get my hands dirty.
Another trick I've used that hasn't worked as well as I'd hope but does work to some extent is, tell others what you are working on, ask them their opinion, and tell them to get on your ass about it. Some friends do this well, most don't, it's not their responsibility to get on your ass, that's your job...but if you have it in the back of your mind that someone might ask about project X, you'd better have updates.
TLDR: multi-tasking at a larger scale.