Keep learning in an AI world

Despite what you might hear about ChatGPT being great at coding, it’s not. Even I, a novice coder, can tell you that.

Over the past few months, I have been learning Cypher coding to create knowledge graphs in Neo4j. I learned Cypher the traditional way: reading the official Neo4j manual, doing many coding exercises, going through an online Udemy course and building my own scenarios.

I tried to use ChatGPT to write Cypher code but I found that it often made mistakes. I could spot the errors immediately because I knew what proper code looked like.

ChatGPT was more useful at explaining if my own code was robust, or suggesting different ways to build relationships between data points. Claude AI was much better than ChatGPT at Cypher, but even then, I had to guide it very carefully using my rudimentary coding knowledge.

This learning reinforced what I have been telling everyone about Gen AI – you have to be really skilled at your work to wield it well. And in my case, I know enough coding to evaluate the output of Gen AI.

This New York Times story sounds bleak:

“In Boston, Dan Pickett, the founder of a boot camp called Launch Academy, decided in May to pause his courses indefinitely because his job placement rates, once as high as 90 percent, had dwindled to below 60 percent… Compared with five years ago, the number of active job postings for software developers has dropped 56 percent, according to data compiled by CompTIA. For inexperienced developers, the plunge is an even worse 67 percent.”

But the article also states:

“The most popular job titles specific to A.I. include “machine-learning engineer” and “artificial intelligence engineer,” according to CompTIA. Some skills listed in these job postings are “deploying and scaling machine-learning models” and “automating large language model training, versioning, monitoring and deployment processes.” You can’t learn that quickly without a math or coding background.”

The bottom line? While AI will turn the world upside down, we must still learn what we need to learn, and figure out where AI to help us to get the job done faster.