Being a good postdoc

I'm at a department now where I don't really have a group I plug cleanly into. There aren't undergrads and gradstudents who I share a PI or research program with. I'm also not here a lot. This has me thinking about what it means to be a good postdoc.  Being a postdoc is already a sort of limbo, both in academic roles, and in real life roles. This is me trying to make sense of this transitory phase. Your comments on the same are welcome.

My department went through an n-year review while I was in grad school. The evaluating committee gathered all the grad students into a room one day, and asked us about our experience.
"Are the post docs in the department a productive part of the community?"
Most of the room hemmed and hawed for a while, so I volunteered that the postdocs werea valuable part of the department social life, organizing pick up football games, or other social activities.
"That's nice, but we wanted to know about their academic contributions."
Someone else, whose advisor had a postdoc, verified that they were academically useful.

The evaluation committee at my graduate institution may not have valued the social and mentoring role of post docs, but I think that it can't be overlooked. Post docs remember what it was like to be a graduate student, and our experience, being more recent, is likely to be more similar to that of the grad students in the department, than those of professors who graduated a decade or more ago. We are more likely to be at a similar phase of life as graduate students, or at least at the "next phase" of life that graduate students are trying to plan navigating. I've seen friends in fields without a tradition of postdocs turn to young tenure track professors for this same role.

I remember asking professors in gradschool what being a postdoc was like, and I got very vague answers or anecdotes about postdocs that they learned a lot from. When I asked postdocs about the same, they told me that it was much like being a graduate student, but with less guidance from the mentor. My advisor, who also had/s a 2-body problem, told me about the fact that he and his wife moved every year for the first 10 years out of grad school, but conversations beyond that were not really possible. It was from the postdocs who came through our department for a year or two that I got  a perspective on what it is like to live such a transitory life.

My last year as a grad student, my advisor had a post doc. He knocked on my door a month before my defense.
"I was wondering if you'd like to speak in the knot tying seminar. We have some empty slots at the end of the term."
I look at my schedule and reluctantly agree. After the talk, he knocks on my door again with some good followup questions. We talk for a few weeks to see if there is a collaboration to be had. There isn't.

The academic role of a postdoc, working the in the main area of interest of a PI, who has lots of graduate students in the same area is relatively clear. I've never really found myself in that position, so I won't speak to it. Readers are welcome to fill in this gap.

What is the academic role of a postdoc who lives in the periphery of the interests of a department? Unlike previous positions, in this position I find that there are lots of people doing things that I am interested in, even though I am not actively involved in any of these areas. It's a little bit like living in an intellectual candy store. But other than showing up at seminars and asking questions, what can I do to be of service to the department?

• Topics courses: Teaching topics courses are a LOT of work. Be aware of this before you take this advice. However, teaching a course in your area of expertise, that is not the specialty of others in the department, is a great way to become one of those post docs that professors later talk about as someone they learned a lot from.
• Organizing in house seminars: This takes less work than teaching a topics course, but often requires having a PI or someone else with some clout in the department to kick some asses to get people to volunteer to talk. You may still end up giving several of the talks, but unlike a topics course, you won't be giving all of them.
• Volunteer to speak: Volunteer early to speak in one of the department's seminars. If you have a PI and a project that you fit in well with, your PI may already ask you to do this. But if you don't, and are hanging around the periphery, this is a good way to broadcast to others what you do. If there is a seminar series aimed at PhD students, give a short talk there as well. Then the students will also have an idea of what your general interests are, and will seek you out if they need help in something you may be an expert in. 

Having done some of these, and watched others do some of these, I need to emphasize that the earlier in ones position at a place one does any of these, the better. If people know the third week you arrive exactly what your expertise is in, the longer they have to think of you as the go-to person when a problem arises.