I've used Elance in the past and it has worked, but I don't want to have to filter through postings determining the legitimacy of various jobs.
I know the best way is through one's personal connections and network, but I'm looking for an online platform with quality jobs.
Thanks!
The single largest barrier to entry for lucrative freelance contracts is ability to sell, and the second largest barrier is development of a reputation. Sites like oDesk remove both barriers, by standardizing and transactionalizing projects and by relying on a superficial reputation system.
The result is a race-to-the-bottom competition that will slash your rates and pit you against gamesmanship from suppliers that can't even deliver competently, but against which you'll have a hard time differentiating.
If you're considering freelancing over your summer vacation, you have more flexibility than almost all other freelancers. You should exploit that by developing your sales ability and your reputation, which will pay off handsomely when you leave school and it comes time to collect and negotiate offers.
One (of many) ways to do this:
1. Start writing articles (for your blog, but don't call it a blog). Start with your preferred technology stack and gradually branch out. Write case-study style about solving actual problems with Node or Angular or Rails or PHP or whatever it is you work best in. Work in specific languages with specific example, but make the focus of your articles solving actual problems and how you set about the process of generating those solutions.
2. Circulate articles on HN and Twitter; participate in related discussions.
3. Make it clear, constantly and regularly, that you're available for projects that are similar to what you've written about.
4. Optionally, build a "Github Resume" of the code that forms the basis of those articles, and link to it.
5. Start asking friends and acquaintances if they know people who have problems that could be solved well with software development; be forward once you find prospective clients, and, after establishing bona fides (for instance, suggest a 15 minute conversation about their problems and actually provide free advice), follow up by asking for the sale; you have nothing to lose --- "no" is a perfectly fine answer.
You'll do better long term executing this plan than you will on oDesk.
What should I be calling it?
A series of articles.
If you haven't got a portfolio, I'd build one. Even a simple application in your target domain. Ideally, make the application for someone else so that you can obtain a reference. I did this by severely cutting my rates and offering a cannot-be-refused scenario to HN, I'd build your iOS app for $200 [1]. Some people may say this isn't correct to start out by gimping yourself, but it worked for me. I got around 15 offers from that post, and took up two clients. The first I completed in a week or so, the next took me a bit longer, but the client then went on to become a staple and was paying standard rates (I charged £20 an hour).
From these applications, it became much easier to sell myself to clients and demand a higher fee. Now I practically have clients when I actively want them (the freelancer HN post is a good resource here). My rate is still £20/hr, which is fairly low by some people's standards, but I'm comfortable with it as it allowed me to live a comfortable life whilst finishing my University degree.
You should probably change this view because the good ol' world just does not operate like this.
FWIW I have looked at eLance and oDesk and the like on more than one occasion and I find the process very tedious. I would much rather go to my Community first (HN, Reddit etc.) because it is, to me, more genuine.
Sadly, anonymous online marketplaces are not the way to get decent-paying, non-crazy people to hire you for software gigs. I know some people have success with Elance or Craigslist, but I recommend looking at your personal network again.
The trick is not to ask people, "Need any software?" No one will ever say yes to this: no one needs "software." They have enough "software" and they hate the software they have. What they need is solutions to problems they may not even know they have.
What works better is watching people work, or asking them about their work. You can use the pretense of wanting to learn how they use software, or whatever. Just watch them work, and look for pain points, bottlenecks, manual-but-automatable steps, etc. Dig a little deeper to understand the problem as they do. Explore ways to address it with software. Offer to build a mockup/do a demo/write up a proposal (1 page description of what you could do and how it would help) and walk through it with your friend's boss.
To borrow a patio11-ism, this is the "Find an Excel workbook that someone edits and emails to someone else; then replace it with software" approach to finding freelance opportunities. And it's an approach I've stumbled into as well.
My personal testimonial to this approach is that my first freelance opportunity accidentally grew out of helping my mother with some work. I wrote a Java app that did some web searches for her and parsed/saved some numbers. Very crude app, yet it saved her hours a day. Then her boss heard about it. 10 years later, that app has grown into a suite of applications that launched my consulting business, earned me probably $50-80k over the years, and has easily made my mother's company millions of dollars.
What types of projects are you looking for? Having a clear idea of what types of projects you're willing / able to handle will help determine which jobs or contracts you're able to complete.