Would love to share my experience since it was pretty different from most of my peers. I contributed to VTK, building tools to magnify invisible motion, based on a CSAIL paper [1].<p>Unlike most people, I hardly had extensive programming background. In fact my core competency was in the image processing (I come from an electrical engineering background). I never considered my programming skills good enough to participate in GSoC. In fact, one of my friends told me about this project a week before the deadline. I was already implementing this algorithm for a course project of mine, though it was in Python. Hence I spent most of my time learning professional programming and ended up successfully completing the project. In the spirit of open source, I have shared my entire GSoC application publicly [2], I hope you find it helpful!<p>[1] <a href="http://people.csail.mit.edu/mrub/evm/" rel="nofollow">http://people.csail.mit.edu/mrub/evm/</a>
[2] <a href="https://medium.com/@ashray.malhotra/google-summer-of-code-gsoc-sample-proposal-10fdff883963" rel="nofollow">https://medium.com/@ashray.malhotra/google-summer-of-code-gs...</a>