TechTales: Fielding a team at Hack MT '19

Jan 28, 2019 at 09:30 am by tchoate

Slideshow
Hack MT

Hack MT was another great event this year. Thanks to all the sponsors, MTSU faculty and volunteers that work together to make this event happen.

There were many strong teams and some amazing products created over the weekend. See summary from MTSUNews and other media sources.

This article is about the unique experience of pitching a product concept and working with a team through-out the weekend event.

About a month ago, I started working toward a product concept that reflected my personal interests and would be likely to generate interest.  Hackathons generally have a pitch session where people with product ideas present their idea and the tech stack they want to use to the participants in hopes of attracting a team.

Although ML (machine learning) was my focus area during graduate school and I follow ML developments in the news, my technical skills in this area were a little rusty, so I spent some time getting up to speed before the event.

I eventually landed on the idea of producing a suite of web-based ML applications using the Python programming language and the popular Tensorflow ML library by Google (now open sourced). 

Members of the team would be encouraged to build their own ML applications, which would be combined and presented as a suite of web apps with a single interface.

I was pleased that the concept attracted a solid team (see slideshow). The team was dubbed the "Robot Overlords," a tongue-in-cheek reference to the simple-minded nature of the project goals compared to human intelligence. 

The final product presented at the Science Fair presentation included hand-written digit recognition, an app that reads in Shakespeare's plays and learns to babble in that style and an app that learns to survive in a simulation of the classic Flappy Bird phone app. 

In addition to the web apps shown in the slideshow, the team had other apps that just could not be polished and integrated in the time available. Other apps included an ML system that learns to detect credit card transaction fraud, an app to predict which teams will make it to the Super Bowl (timely) and an ML system for facial recognition/recall.

Overall, it was a great experience that I would recommend to anyone that enjoys tech and is interested in working with a team of smart, enthusiastic young people in pursuit of new product/tech ideas.

My thanks to the team members: Zack Toupe, "Jimmy," Ethan Basham, Aiden Murphy, Braiden King, Robert Smith, Abdinadib Ali, Luis Chunga, Harrison Welch, Michael Ketzner, "Colin" and Jerimiah Jayne. 

It was great fun.



Comments

or Register to post a comment

Get the Newsletter!