What is expected of a postdoc?

You'll very quickly learn that being an academic involves more than just writing research papers. Your time as a postdoc is when you can start learning about these other aspects, while building your own profile.

A postdoc needs to do the following:

• Build a publication record. This will involve both what you are paid to do and your own line of research.
• Get involved with supervising students. Help with the PhDs in the lab, and get involved in supervising masters students.
• Get involved with obtaining funding. This can either be by helping your employer or (ideally) obtaining your own funding.
• Build an international reputation.
• Start collaborating with external parties.
• Gain some teaching experience. This is absolutely crucial if you want a faculty position.
• Learn how to manage projects and a lab. This includes overseeing the progress of projects, allocating your time (and others), presenting results at meetings and writing deliverables. If you are in a lab setting, you will need to learn how to fix/calibrate/maintain critical equipment and software so that you can start your own lab some day, and you will need to become proficient in teaching more junior members on how to use that equipment.
• Start to devise a strategic research plan. While it is fun to do opportunistic research, solving a problem that comes along or investigating any idea that pops into your head, a better long term strategy is to formulate an interesting long term research plan and follow it, building result upon result.

Be as proactive as humanly possible, without being annoying. Talk to everyone in the department, especially people whose research interests are close to your. Go to conferences and sit down and work with interesting people (not necessarily the superstars).

Let me add one item that Dave Clarke omitted, which I think is actually the most important:

• Separate your research reputation from your advisor's. Congratulations! You have enough of an independent research record to land a postdoctoral position. Unfortunately, that reputation is almost certainly deeply entangled with your PhD advisor's; deep down, many people in your research community still wonder if (or simply assume that) you've just been riding your advisor's coattails. Your primary job is to convince them otherwise. Do not work with your advisor, and do not work in the same subsubsubfield as your PhD thesis. Make a name for yourself as a truly independent researcher and scholar.

And a secondary corollary:

• Do not just ride your supervisor's coattails.