Breaking Down a Tech Talk

Breaking Down a Tech Talk


Chris Meyer

Joel Nishanth Ponukumatla


Public speaking has never been a strong point for me.  In fact, it can be mentally crippling for me to even attempt.  Recently, my team finished a rather significant project and was asked to give some presentations on the work we did, however, between speaking to a very large group of people and getting questions I wasn’t prepared for I started slipping badly.  I would get defensive and struggle to articulate what I was trying to say. I realized this problem of mine could be affecting my credibility as an Engineer. I decided enough was enough and I would put some real effort into becoming a better speaker.


A week before KubeCon, Nóva put out a tweet offering some of her time for a one-on-one with whomever might need it.  I figured “Why not?” – I could ask some questions about how Kubernetes is managed on a large enterprise level and maybe try to get some public speaking tips.  We met in a hotel lobby on day two of KubeCon, and she showed me this really cool diagram of how she structures her talks. She went over each point of the diagram and how it flows:

Nóva’s diagram

After the conference I wanted to make sure I had a reasonably good handle on the information I was given, so I started watching Nóva’s Falco talk, mapped specific moments in the talk to states in the diagram, as well as some details within them that Nóva pointed out. Going through this exercise helped me to understand how some of these conference talks are structured and how to go about putting one together.

For me, the biggest takeaway in all this is that talks are just good storytelling. 

Listening to the Falco talk you see that it’s a story about runtime security in Kubernetes using Falco, where different options of monitoring the kernel are explored before ultimately leading to using eBPF, and why it is the right choice.  The following sections relate to each state in the diagram, and my takeaways from chatting with Nóva:


Establish Credibility & Bitly Link

The introduction of your talk is a good place to try and tell a joke or two to keep things light.  It’s also where you want to ask questions to try and gauge who your audience is and figure out at what level you should be speaking.  Asking questions can give you a moment to take a breath and collect yourself if you start to feel overwhelmed.  In other words, questions are good for gauging your audience and giving yourself a break.

Check out her 2019 FOSDEM video here. She does a great job of introducing herself and why people should be interested. Then she tells a joke to give herself a moment.


You’ll also want to establish your credibility as a speaker.  The audience should know that the person they’re listening to knows what they’re talking about.  It’s a chance for you to tell the audience why they should be listening to you to begin with. Typically this might sound like:


“My name is … I’m a … with …  I work on such and such project(s)”


It’s also good to offer your listeners a link to more content about your topic.  Bitly is useful because it allows you to gauge how many people are following up on the topic you presented.  You should have these link(s) available at the beginning and the end of your talk. Additionally, before the end of your introduction, it’s not a bad idea to briefly go over what you’re going to be talking about as well.

Problem and Constraints with Buzzwords

It is important to keep the overall goal in the picture, once you’ve established your credibility you should ease into the problem and explain what it means to you, i.e., what are the constraints you are facing with this problem? Make use of “buzzwords”  as that will enable us to keep the talk relevant to the audience and capture their attention. It is important to establish the critical core of the problem.  

Example : We are facing an issue with security in Kubernetes , what does security mean in the kubernetes realm? (Here we talk about what it means and drop a hint of your solution without going deep into it.)  Why is it crucial? How did Google solve it? Google solved it by using a really expensive mechanism which cannot apply to us, as it is not frugal.

Short 1,2,3

Typically, a speaker will have 3 to 5 talking points to support the call to action (the diagram uses three).  To help guide the listener through your thought process, you identify the potential actors of your solution and take a breadth first approach when talking through each of your points via a quick overview.  Later on, you’ll go deep into the roots of your points and continue until the whole approach to your call to action is covered.


Short version of your solution which consists of products/solutions A, B, and C: 


Example : If your solution consists of products A, B, and C.  Make sure to call them out first as you will be referring to these actors throughout your talk.


In the Video we have the speaker breaking into the solution, and she has 3 actors: 


You should already have the potential facts you want the audience to focus on prioritized.  We can also see that the three introductions blend in with the beginnings and endings, so she was able to create a story from the actors which are tightly integrated with each other. 

Long 1,2,3

In this section, you want to start digging into each of your points with more detail.  The speaker has slowly leaned into the product, she starts out with an introduction and history of Falco, then deep dives into the solution.  She slowly leans into kernel module from which it sends a signal to the ring buffer and eventually lands into Falco in user space. She then goes on to talk about the components of Falco and how it works. 



Run through your points again while your audience is engaged in your solution.


Example : Let’s assume I am solving a math problem, I use formulas A, B, and C . I want my audience to focus on my main emphasis on formula B, so I run through the solution using formulas A, B, and C, then go back to formula B and repeat what formula B is all about. 

Problem and Constraints with CFP

This is where you restate the problem and constraints while incorporating your CFP (Call for Papers). You’ve walked the listener through the thought process of how you came to your conclusion and now you want to remind them of what it is you’re trying to solve.  You’re setting yourself up for your call to action so this is where some drama comes into play to remind the listener why this problem is important.

Action & Bitly link

This is the conclusion or, simply put, this is when you start to dig in to what you want the audience to take away from your talk, be it a solution to a problem, or just a better understanding of your point of view on a topic.  The call to action can be supported by a demo when appropriate.  This is also a good time to give your audience another chance to grab your bitly link or any other resources you want to refer them to.  


So, in theory, you’re done with your talk now.  This is where you thank the audience for their time and take questions.  Questions from the audience can help to point out something you forgot to cover or expand on more in a future talk.


In Closing


Hopefully, this post can help some people out.  To me, listening to a talk is kind of like watching a movie where you get lost in the story and never stop to consider all the details of how it was put together to begin with.  I’m scheduled to do another talk for work in a few weeks, so hopefully with a lot of practice and a guide to follow I’ll be better prepared to deal with the pressure and handle myself a little better than I did before.  Wish me luck!


0 comment