The journey so far is a little stochastic. With this project, I’m embracing my natural project management style. Of course through this entire program we’ve been taught many frameworks for project management, but to me, if they don’t come naturally then they aren’t optimized. This project breaks a lot of rules- by using ChatGPT, we throw away all preconceived ideas of how a project is supposed to look. We start with a bare-bones working app and progressively slap it into place until we have a working product. This is familiar to me because through my career, I’ve always approached projects similarly. I do a burst of work and then evaluate, and do another burst. I understand that it by no means works for everyone, but it’s what’s worked for me in my career so far, and now with advent of these new coding assistant tools, it’s easier than ever to work ‘explosively’. I don’t worry about commits or passing set requirements- I make something that works and then I try again, making it better. This project is also unique as a team project in that we work independently- naturally, if we were collaborating for more than the theory level of this project, I’d have to compromise how I approach projects.
I feel like my project is on-track to succeed in that I have learned a lot about how to use ChatGPT as a programming tool and have recommendations to make to programmers on the future who want to optimize its usage. It is difficult however to precisely shape the code in a desired way. Interacting with all these AI tools to me feels very dreamlike: asking ChatGPT to create part of the project and then shape it accordingly is like building a sandcastle in your dream. It’s very nebulous and difficult to wrangle, if that makes any sense. As a research project, however, this kind of observation is precisely our goal.
This project is going fine. It’s not going traditionally, but that’s part of why I picked it and what I’m enjoying about it. I’m really looking forward to seeing where it will take me next!
Leave a Reply