[cweagans]

Just build something - it really doesn't matter what. You have a good window of time where you have very little requirements in terms of paying rent, bills, etc., so take advantage of that and just write as much code as you possibly can.

[jc4p]

Bingo. This is exactly what I did. Build a lot of of things, it doesn't have to be unique, it doesn't have to have a business plan, it just needs to be something.

Build tools to help automate simple things you do every day. Build tools just for yourself. That way in a couple years you'll already come to the field with a huge portfolio, and who knows one of your random projects might actually end up getting popular and making you something.

I started making personal projects at 15, worked under my mom's name for freelancing sites while <18 (illegal against their ToS but oh well) and now I'm leading a team of people at a big company since I have 9 years of experience and 5 years of experience with my specialized field (Android).

Just build something.

[anon3_]

When you did that, was it a website? A service?

Simply having an open source project in itself brings in no cash. Can either of you (or someone else) go into detail on your experience? Links are OK! The OP (and myself) would love inspiration :)

[xxtjaxx]

Had experience in sysops perl scripting for web apps under my belt (wanted unified dashboard + designer for monitoring metrics visualization https://github.com/andrdeas-marschke/nagplot) in the process reverse enginieered NRPE from source / wiretracing and created one of the most if no the most linked articles on the protocol in the process. Released the code as seperate package (Nagios::NRPE).

Then decided to learn JS and node and started writing a web real user monitoring system based on http://github.com/bluesmoon/boomerang. While at it submitted PRs to upstream and engaged in the community mostly with complaints about sftandard stuff missing and modules not being in a good state.

Now I work in the same company as the guy who created boomerang and lpve every minute of it.

[cweagans]

Also, I contributed to Drupal for a while starting in high school. That itself doesn't bring in cash, but it established a reputation in that community, and now I work at NBC doing Drupal development. Open source contribution is a means to an end sometimes.

[cweagans]

I built simple websites for local businesses, and then moved into building an overly complicated CMS and hosting service for them. I put in like 18 hour days for almost a year (skipped a lot of classes in high school), and ended up with a pretty cool product. Unfortunately, I lost pretty much all of the code, so I don't have a lot to show for it. :(

[arielm]

I started by building services but never really expected a payment, I just wanted to build something and figure out my way around servers. I think that's the best way to get started because it affords you the opportunity to start over if you find a new framework/database/server/etc. and there's no penalty for failing.

[Lambdanaut]

I built a social website for artists. It exposed me to the whole stack and I learned an immense amount. I got my first paying gig by using the website I built as a foot in the door. Nothing beats learning by doing.

[Pfhreak]

I'm a dev who became a manager. I do a lot of interviews at one of the big tech companies. I cannot overstate how right cweagans is -- you have the opportunity now to just build stuff. Explore new projects, get passionate, learn as much as you can. You don't have to finish every project, but try to get one polished and shipped. The first question I often ask candidates is to tell me about a project they found challenging and interesting, and if you have a wall of those projects you'll be a leg up on your peers.