How I think devs should and should not approach AI in 2024

First off, Happy New Year! You made it. 2023 came with its challenges. The IT job market was a bit of a whirlwind. Companies pushed for (or in some cases imposed) a return to work (in-office) policy for employees. In the midst of that, the (arguably) biggest buzz of the year was AI and the battle for supremacy between different providers. It's become pretty clear that ignoring this buzz, as a dev, might be akin to an Ostrich burying its head in the sand. At the same time, careful consideration needs to be given to any enthusiasm for it. So whats the way to go?

DEVS SHOULD

Get comfortable with using AI as an assistant. Taking a look at some of the tools out there, from OpenAI's ChatGPT, to GitHubs Co-Pilot, to Twitters Grok etc. it is immediately apparent that the current offerings have been tuned specifically to boost productivity, save time and provide quick access to information. Obviously the other GPT models available to devs from OpenAI allows for a chance to build or improve an existing app, but if you don't have an immediate need for those, its totally fine. However, incorporating an assistant (whichever one you choose) into your day-to-day dev workflow, will give you an undeniable edge.

DEVS SHOULD NOT

Worry much about creating their own AI models from scratch. While it's ok to have a high-level idea of how the models are created, without a pre-existing education/knowledge and love of Math and Statistics, its probably best to leave the model creation to the PhD's. Trying to achieve success at building your own model(s) is like being a great race car driver who suddenly wants to get great at building a Formula 1 race car. Its possible, but hard. Instead, devs should play to their strength (development experience) and let that align with a helper/assistant as I described above.

Have an awesome 2024. I'll see you on the other side :)