I just gave my first academic talk at a conference at SJTU IIC yesterday. I presented an ongoing project, and I had hesitated for a long time about whether to give a talk on an “incomplete” work. It turns out much of my nervousness was unnecessary—I received a lot of helpful feedback from the audience. So I’m writing this post to summarize some lessons I’ve learned during this journey. Many of them echo the insights from Paths to Research by Christopher Ryan and Runshan Fu (which I highly recommend!): Paths to Research (PDF)
Be confident
In my first draft of the slides, I presented the results with a negative tone. I said things like “we use this strong assumption” or “this technique might be helpful.” I did this because I was afraid that some smart audience member would immediately identify the limitations and criticize my work, so I tried to preempt them by pointing out the weaknesses myself.
It was my collaborator who reminded me to be confident—especially when presenting to others. If your entire talk focuses on what doesn’t work, the audience might wonder why they’re even there. Even if someone in the room is an expert, they haven’t spent 30+ hours a week thinking about your specific research question. You are the person who knows the most about your work, and that alone is reason enough to speak with confidence.
On the purpose of a talk
Here’s my personal understanding of what a talk is for:
- It’s an advertisement for your work. You’re not there to explain every technical detail. Your goal is to highlight the most compelling parts—motivation and key results—and spark enough interest that the audience will look up your paper afterward.
- It’s a chance to collect feedback. In this sense, presenting an ongoing or incomplete project is actually better. Since it’s not published yet, you can still revise it freely. I got many valuable suggestions this way—it’s honestly one of the most efficient forms of peer review, better than sending a draft to other researchers.
- It’s an opportunity to connect with people. Beyond the Q&A, audience members might approach you afterward to talk. One PhD student came up to me and said how happy she was that she finally understood an optimization talk—she works in Operations Management and usually gets lost. Her words reminded me of the importance of making my talk accessible. If someone gets lost early on, they may spend the rest of the talk feeling uncomfortable or even doubting themselves.
Prepare for questions
I didn’t prepare for the Q&A at all. I figured, “It’s the last talk of the conference—who’s going to ask questions?” Big mistake. I got so nervous that I didn’t fully understand the question and gave an off-topic answer. That moment cost me the chance to explain my work more clearly and in greater depth.
Luckily, I later found the questioners and had some great follow-up conversations. But I learned the hard way that anticipating possible questions—and rehearsing your answers—is absolutely worth the effort.
Finally, I’m deeply grateful to my collaborator for helping me rehearse, and to everyone in the audience for being there and offering feedback. Their encouragement gave me a big boost of confidence for my future research journey.
