Tuesday, April 2, 2013

Leadership by Stefan Posey

The following tips/ words of wisdom are from the past president of the Georgia Tech RoboJackets  Stefan Posey (stefan.posey@gmail.com) 

Time, Sacrifice, Effort & Who works for who

I never ask others to do more than me, to put in more time, to put in more effort, or make bigger sacrifices. You will always work for the people below you as they depend on you to achieve their goals and meet their milestones. They don’t work for you. They work for themselves and you work for them. This is true for both volunteer and paid jobs. Quality people can and will, if needed, find other ways to do what they need to do. The ones that step up and try to match you in at least one of the ways (time, effort, or sacrifice) those will be your core of dependability. To keep these people you must remember the above as they can find another source of compensation should they feel inclined to do so.

To Quote The Game “I need some meditation so I can lead my people”

Being the workaholic that I am I really suck at this. But its true you do need it. I like rap/hip hop (ATL stuff & pac), quality R&B, big cars with relaxing rides, and a decent movie/tv episode. Also hanging with people that don’t constantly need help and can think on their own. Watch out for those under you. Understanding what's going on outside of the project can give you perspective and help you enable your team to be more effective. You will be largely surprised by what's up behind the scenes. When people are pushed hard they can and will break in unpredictable ways. You are not an exception to this. Be aware as the definition of hard is variable.

Be Prepared

Thats the best lesson from scouting. Yes, I’m infact an Eagle Scout. People may ask why on a project do you have spares of X, extra cash, and they won't understand. Some will even say that having extra of Y “just because something might happen” or there are unknowns is retarded. Don’t listen to these people. I keep a bottle of motor oil, trans fluid, a rubber mallet, jumper cables, and some type of canned food in my car. Those items aren't always for me.


Money does one thing and one thing only. It expedites the resolution of easier problems. Its a tool never forget that. Always focus on the cash I/O when trying to diagnose situations. Meaning you need X, how much are you burning, how much do you have, and how much are you keeping in reserves. The proper reserve amount is really dependent on what you're doing you’ll find this out after a few disasters. A good starting point to try and estimate what you might need as backup would be to look at the problem or task that is the most critical that has the most unknowns. Then looking at the cost of reattempting failures in this path.


The best ones always present themselves in strange and non typical circumstances. I once showed up to a 1.5 hour midterm 45 min late. Walked to the front of the class requested a test from the professor and was handed one. Apparently that was the ballsiest thing they'd ever seen. The end results weren’t great, but I did do pretty good on the second half which the entire class failed. Doing that on 2hrs of sleep wasn’t going to be spectacular anyways. Ended up spending the night cleaning up shop for a suprise safety inspection as a few days before the the high school group had shipped their bot and the shop was pretty rough. The point is sometimes things that might seem overambitious are a lot simpler than you initially think. Also you will have to fly from the seat of your pants so get comfortable with it. Acting without perfect info is often mandatory.

Soft Power and Helping People Out

Personal favors are worth more than you think. True helping someone out is a bet, but its a bet thats paid off more than once for me. An example I will share: Often when running my team in our shop people would drop by asking for help or advice on a project. One day a girl showed up stating that she had heard this could be the place to get some wood cut for a project for a class. I didn’t really have time, but it was simple and fast (ended up being late to a meeting). Sort of forgot about this girl for a bit because it was so random and simple (for me). Fast forward a few years this person is highly involved with ANAK and tries to help us out when our shop is getting torn down. Quality people will do what they can to help you out in the future and those that don’t won't be worth your time later on.

Chains of Command & Asking Questions

All chains of command are doomed to failure. This can be both good or bad depending on who and where you are in it. Perpetually sitting at the top can lead to blindness of this. You should try to jump in occasionally when you can if only for diagnostic reasons. When given a path to create information flow that covers many links do it. Just because people above or below you have responsibilities doesn’t mean they are fulfilling them or understanding them. Find out because you need them to be functional so you can be effective. To do so I routinely will ask multiple people the same / similar questions. Also the same person variations on the same question. Is there miscommunication, is the communication slow, which direction is the speed and from which point do things travel from the slowest and the fastest?


All players should know everything (or as much as possible) about the objectives. My approach and what I think works is email lists. That way team / subteam X knows where team Y is. I keep my email bulleted and in task form. If I need someone to do X the task lists starts with their name. Everyone should get everyones task list as people will need to collaborate. No one has time to read through paragraphs to figure out what you need them to do and they might miss something or interpret it wrong.

Small Details

These can be very important for overall implementation. There are people that spend allot of time getting that plastic bezel to feel just right on that mobile device you hold in your hand and make it happen the exact same way thousands of times. Its true small details can make or break any project or product. However small details should not hold up an entire projects progress. I call this the “tables and chairs” problem. If you're running an event and your biggest concern is where do those go and what color your balloons will be your either really close to done or impossibly screwed. From what I’ve seen it's often the latter.

Situational Awareness & The Critical Path

I would always make people plan out what they need to do, what they need to do it (tools, people, money), and define the minimum required functionality. Your first step should be to see how those under you see the problem. An example could be “Man I’m really having problems implementing sensor X, but the robot can’t yet drive and competition is in 1.5 months in a foreign country that doesn't speak your language.”  Next up,

  • What is your critical path? You might have more than one.
    • Sometimes you can redefine these to alter their dependencies and as a result decrease some bottleneck in project flow.
    • I like to simply define a critical path as relatively long serial steps that can’t be broken down into parallelized actions.

  • You need to figure out what resources do they need?
  • Why aren’t these resources or tools currently available?
  • Is this for test or final product?
The last one is an interesting one as for testing and validation you might have allot of substitution possibilities. In the end the robot needed to be able to drive first and the degree of  implementation of the sensor was scaled back.

Overoptimism, Being a Pessimist, & Measuring Success

I’m pretty much a everything is always screwed and will be so we need to man up and deal with it type of person. Sometime I get a bit of static from others due to this. However it is a quality of mine that has allowed me to look at situations and try to anticipate possible failure paths. Be wary of blindly over optimistic people they can cause strange problems to occur. In the end success is what you make of it to yourself. However, how others see it can make for quality feedback.

1 comment: