Ask HN: I have an idea, but need help. Where do I go for it?

2 points by wakkaflokka ↗ HN
I've had a project idea that I've wanted to do for a while now. Unfortunately, I don't have time or know-how (on the programming side). I've tried to start the project in Python, but always realize it's going to take more time than my wrists can handle (I work as a data scientist, already too much time at the keyboard).

I'd like to create software that will take all your personal data that you are able to download - like Google, Amazon, Facebook, LinkedIn, Mint, Health Data, SMS/texts, RescueTime, RSS Readers, CVS prescription history, 23andMe, etc. - and compile it into a "personal analytics" software. For example, help answer questions like:

1. Where do I spent the most money around town (location data + mint transactions)?

2. Where do I send the most angry texts (location data + NLP on texts)?

3. At what time and after what Google searches do I end up buying the most junk off Amazon (Google + Amazon)?

4. Recreate your life as a movie (Google street view + texts + purchases + etc.).

Those are just the start.

My first foray into this was writing some basic Python code to see where my girlfriend and I crossed paths before we met (we were within a mile of each other at the same time, two years before we met).

1. This has to be all done locally, no cloud involved because of the clear privacy implications. What kind of local DB that interfaces well with Pandas/Python would be appropriate?

2. How would I encrypt the data locally, and potentially allow 'agents' to interface (for example, seeing where you overlap with somebody else in location)?

So the Ask HN question is this - when you have an idea and want it to materialize but you need help - where do you go?

If you have a little bit of money, do you hire a freelance programmer?

Do you post a barrage of Stack Overflow questions?

Do you partner with somebody, or do you throw the idea out there in hopes somebody else might pick it up?

2 comments

[ 3.2 ms ] story [ 14.1 ms ] thread
"If you have a little bit of money, do you hire a freelance programmer?"

Long before you spend any money trying to build this, you should probably do some research to see whether there would be any market for such a product.

For example, see if you can find 50 people (average folks, not data scientists) who think this might be something they'd want to use.

Personally, I can't think of any uses I'd have for such a product.