Ask HN: How do I make my PhD defense presentation great? | Hacker News Reader

Ask HN: How do I make my PhD defense presentation great?

2 points007024 comments

I will be defending my electrical engineering PhD (in the USA) dissertation soon and wanted to ask the wiser folk in here for some advice on how to have a great and memorable (in a good way) presentation. The presentation part of the defense will be about an hour long and my field is robotics.

I'm fairly comfortable with presentations and know the usual rules about not reading off the slides, not having walls of text on the slides, etc. Usually my presentations are very visual with minimal text and I like to keep the focus on me instead of the slides, this allows me to control timing. I also almost always bring props (PCBs, protoypes, etc) related to what I'm presenting on.

With all that said, I've never defended at the PhD level before so I'd love to get some pointers from you all on what I can do to go that extra mile and make the defense great.

View on HN

A defense presentation is very different to one you would deliver to a seminar audience. The panel want to confirm that you have made a significant contribution to the field. They know the field and are unlikely to be swayed by video of robots doing their tricks.

A good format might be to present a 5 minute segment to provide an overview of what you have worked on. Then the remainder should focus on reinforcing the contributions that your dissertation also presents. Departing too far from your dissertation might raise more questions which you will then need to defuse during the interrogation.

Obviously if your contribution includes something that can be further evidenced by a suitable short video segment, then that will be valuable. For example, if you have developed some algorithm which leads to faster, smoother, etc operations under some adverse condition, then demonstrating it in addition to explaining the theory behind it is of value.

BTW, most of us only ever have to defend one PhD.

Yes. Provides a 'double accounting' approach to getting comfortable outsize zone to build confidence in in-zone knowledge.

Assuming suitable material /subject area in which it's reasonable/feasible to do.

1) Non-technical audience -- source/feedback for relevant presentation ideas/concepts.

2) technical audience that can judge how well non-technical presentation correlates to presentation of thesis material.

Other than algorithms :

https://news.mit.edu/2022/robotic-actuator-fly-0621

https://paperpile.com/g/thesis-defense/

Stocastic or non-stocastic presentation?

Use illusionist / marketing survey tricks other than that of the defense committe kind.

pre-seed audience with relevent topic area questions/titles before defense of the 'poor phd student' kind while

being mindful of thesis defense committe bounds.

Practice with audience outside field with people in field observing. If can make sense in reasonable amount of time for non-field audience AND person(s) within field -- good to go.

For taking the ego down a notch -- explain/give talk to middle schoolers/high schoolers. If they're hooked & makes sense to in-field observer, very good chance good to go.

'double accounting' method feedback (given enough lead time) can also provide practice ideas/source of potential questions (non-technical & 'how would someone with defense committee ask/approach this?)'

Reply on news.ycombinator.com