So, I have had a year out to work and am now going to do my Master's degree in Computing. I have decided to fully embark on a career in academia.
What can I start doing NOW that will help improve my academic reputation, credentials, CV and career prospects? For example, should I do more writing, set up a technical blog, find research opportunities, what projects should I start, organisations to join etc.?
Since you are early in your career, I'd actually advise against pushing too much on specifics of reputation building. That can come later, but you can afford do let it develop naturally.
Instead I'd advise two things. The first is to learn you subject as broadly as you can and to pick one or (preferably) two sub-topics that you are interested in and into which you can delve deeply. This might depend on local resources, of course. Eventually you will need to become the world authority on some (perhaps tiny) area, but getting there depends on knowing a lot of related things.
The second is to find ways to associate yourself with researchers and scholars, whether as a member of a research group (as others here advise) or just finding a mentor or two whom you would like to emulate. Read their papers, ask for ways to delve deeply, etc. See if they will listen to your ideas. Ask them to point you in profitable directions.
But don't neglect your studies.
One way to associate with other professionals is to attend conferences in which the superstars not only present their own work, but are willing to sit around during coffee breaks and discuss ideas in the field. Even if you don't contribute to such discussions, listening will teach you some things, though sometimes they just discuss dogs and pubs (not publications), etc. Even better if you can participate in a conference as a presenter of some kind, even with a poster session.
Learn to swim in a sea of ideas. Get feedback on your own.
Writing is always good. It is especially good to write in a journal in which you date your ideas and get someone else to sign and date your entries. If you have especially good ideas this can establish your priority. Writing for publication is a more difficult skill. It takes practice, and feedback. Reading is just as important, especially for a new comer. You need to know what has been done so that you can build on it.
Try to find active researchers who are willing to let you do a research project with them (easier said than done, but if you're offering to work for free in your spare time without huge amounts of hand-holding then people will sometimes give you a chance), and then collaborate with them to write a paper. Papers, particularly high-quality papers in good venues, are the primary currency in academia. Writing a technical blog might be fun, but realistically won't affect your academic reputation one way or the other (it's unlikely that many academics will see or read it). Similarly, joining organisations isn't on its own going to have much impact on your academic reputation - people generally don't care if you're a member of e.g. IEEE, they just care if you've done any useful work lately. Helping organise seminars that are popular in the field (if you can find a way to get involved in that) might get you noticed to some extent, but only counts a bit (people hiring researchers don't want to know whether you can organise talks, they want to know if you can produce interesting research).
Basically, the main thing that matters is (good) papers - try to find ways to write them, preferably with other people, but on your own if necessary (the latter is hard and less effective, so try to avoid it if you can). Try not to waste too much time on things that consume your energy but for which you don't get any reward. Enjoying them can count as a reward, incidentally - by all means do fun things just because you like them, just don't expect them to have a huge impact on your reputation.
Source: It's Sunday, and I'm writing a paper, so at the very least I believe the advice I'm giving :)
I wanted to highlight issues that would be helpful especially in the beginning of a career, but somehow went over to the general character traits.
And finally, you can just try it. It's fun! Well, mostly. Basically, everything up to (and including) PhD is a free trial phase. From postdoc onward the rocky road starts.
