Giving a Good (Academic) Job Talk

I have attended and 'judged' a lot of job talks, mostly in computer science for roles at all levels, increasingly now across all the engineering disciplines. Some good, but too many bad. For the most part the rules of giving a good job talk are the same as giving a good presentation generally. There are plenty of sources for this and I will therefore confine myself to the some of the concerns specific to academic job talks.

Before we start we should split job talks into three distinct types: standard seminars (about an hour, with questions, to a larger audience drawn from the Department), an abbreviated seminar (perhaps 30 minutes, commonly to the panel perhaps with a few additional attendees, with limited questions), a short talk (8 minutes or so, to the panel, with questions rolled into the interview). Each type of talk requires different strategy and skills. Each is used to reflect a different situation and perhaps a different set of requirements on the part of the interview panel. In many ways, the shorter the talk, the more challenging the task. The 'elevator pitch' short talk requires at least as much preparation as the seminar.

In each case you will be expected to be thoroughly prepared, have excellent materials to support your task (this may not be required in the shorter pre-interview pitch), talk with confidence and answer questions with authority. Because job talks are often taking place within strict time constraints, honour these. Do not overrun.

It may appear invidious to remind candidates to 'read the exam question' before they prepare a talk. You will have probably received an email with details both of timing and expectations, read it carefully. Not only will it give the essentials, timing, duration, audience, it may give further clues: if you are asked to present 'a vision for your research at university X' or 'an account of your research' or 'a presentation of a research contribution', then do precisely what you have been asked to do. Do not simply reuse slides blindly from a previous talk, and certainly do not leave clues in your slides that suggest this is what you are doing. The brief will probably contain clues as to the main recruitment criteria, be sensitive to them.

First and foremost be aware that in any job talk you are addressing a mixed audience. There will be experts, but there are likely to be attendees from across the discipline and beyond. This is not therefore, a standard technical talk addressed purely to the cognoscenti. You will need to explain and motivate your primary interest and lead people through the work but, and this is important, do not dumb it down. You should show you have technical depth and that your work is demanding. Getting this delicate mixture right is critical: test it on a range of people to ensure you strike the right balance. You may want to confine some detail to 'hidden slides' available to you in the event that questioners want to follow up on matters of particular detail.

You will be judged, possibly harshly, on your personal organisation, ability to handle the AV facilities without a fuss, audibility, graphic quality and legibility of your slides, and other matters that in a normal academic setting would be overlooked. Remember that the panel is looking for somebody who can represent the institution and who can, with confidence, stand up and teach a class of students. Rehearsal is essential and you should seek out an opportunity to try out your talk with a small, highly critical, audience. It is always good to have a set of 'note-sized' printouts of slides to give out, but I suggest only distributing them after your talk so that you can keep control of the pace and the way the material is 'revealed'. This adds an extra touch of professionalism.

Feel confident in your materials. This is probably not the moment to be too adventurous. It is not, for example, the place to try your first 'prezi' or a rocky live demo.

You will be expected to make reference to activity at the place you are interviewing. Make precise and relevant links that show you know how your work relates to, or complements work ongoing in that institution. Even if the speaking brief is explicitly research focussed you should draw attention to the pedagogic and knowledge transfer dimensions of the work.

Link your work to your CV. If you are talking about published work ensure you have signalled the papers in which the work appears. Be careful with crediting work and images appropriately. Attendees will be on the lookout for scrupulousness in this regard on the part of a potential colleague and collaborator.

Because, academic posts are often held for a lengthy period you need to show that you get the big picture and are addressing major problems that have longevity. This is a rare opportunity to convey the larger vision and to signal your ambition.You also need to showcase your skills, that is what are you 'bringing to the party' in terms of technical capabilities. Signal your individual technical contribution to joint work clearly.

Breadth and scope of contribution are important but be sure to make your talk more than a catalogue. Choose a piece of work for a more detailed exposition. If you can argue it is representative this is good.

Listen particularly hard to questions and answer them patiently and politely. Try hard to be open and responsive, do not reel off stock answers. Questions in job talks tend to be more pointed than in a 'normal' talk, so try to work out what the questioner is seeking to elicit. Are they, for example, testing your technical understanding, your critical faculty or your willingness to acknowledge limitations? Of course, it may be that they are simply trying to clarify something from your talk. Obviously, treat all questioners whether on the panel, or not, equally.

If asked to give one of the two varieties of short talk think out the content carefully. The biggest risk for the mid-length talk is trying to pack too much in. I suggest ruthless self-editing. Keep your slides to a minimum, have a strong clear and well shaped narrative. Leave time for questions. In the case of a pre-interview pitch you should avoid using visual aids altogether, perhaps a single diagram in printed form only. Messing with AV will take too much time and will be disruptive to building a quick relationship with the panel. I suggest sitting at the interview table to present. Use personal notes. Do not apologise, do not give a lengthy preamble, have a few sharp points to make supported, ideally, by nicely chosen quantitative data. Stick absolutely to the brief and do not extemporise. In a short period you have to be 'true to yourself', do not waste time saying what you think the panel wants to hear, this is likely to be transparent and means that you lose the small element of control you could gain over the shape of the forthcoming interview. Conclude by asking the panel whether there are any points they wish clarified.

Above all, remember that all interviews and job talks are as much about you selecting an employer, as an employer selecting you. Smart, highly engaged, students are a very good sign of the sort of place you might want to work at. Vigorous, informed questioning is a signal of a strong academic culture. Rudeness, aggression or competitive questioning are negative indicators. Rule out anywhere with vile coffee.