Struggles, Fun and Code
Staring at thousands of lines of code you have never seen before and being told to change it can be a daunting task. However, that is the goal of Windward’s Code Wars. Teams of college students are given code they have never seen before and have 8 hours to edit the code to make their A.I. program outperform the other teams’ A.I.
As an intern of Windward, I was given the opportunity to compete against 4 fellow interns in a code war to test the 2016 Windward Code Wars clients.
This year’s code wars modeled the board game…
Whoa! The editor just censored the section I wrote here because we don’t want to give away any hints to this weekend’s contest. You’ll just have to trust me that the game is really cool.
Right out of the gate I had some issues with my local computer, which didn’t allow me to run the C++ and Java clients (the two languages I know best). I decided to use the C# client, a language I have rarely used. I thought that my chances of doing well were thrown out the window. (Turned out I was wrong.)
So I spent the first few hours not worrying about coding and focused on learning the code and the game. I ran through the code I was presented with so I had a solid understanding of what each function and class did.
I also used the web to learn strategies for the game and strengthen my C# knowledge. After that it was game time. I went to work on my A.I. making changes and testing often to see what worked and what didn’t.
At the end of the day, all the interns went to Dave’s office to test our AI’s against each other. We ran the game 10 times and whoever
…whoa more censorship here!… after 10 games was the deemed winner. I started off strong by winning the first 2 games, which gave me a solid lead. I maintained the lead for 8 rounds until one of the other interns swooped ahead for the last-minute victory.
This was a great experience and I learned a lot from it, including these 2 key things:
- One of the most beneficial tactics is testing often. Whenever you implement a new feature, test it a few times to make sure it works and benefits you. There were a few times that I thought an algorithm would be super effective but it ended up not working at all.
- Focus on one aspect of the game at a time. If you try to implement
censored, censored and censoredAI simultaneously, it will be very overwhelming. Focus on what you think is most important first and go from there.
I hope this was beneficial for you! Good luck and may the code be with you!
P.S. If you haven’t signed up yet to participate, there’s still time at the Code Wars 2016 registration page.
Author: Connor Shore
Connor is a software development intern at Windward and a student at CU Boulder. He enjoys soccer, skiing, music, programming and outdoor activities.
Other posts by Connor Shore