Many companies are excited about generative AI but struggling to find their best use cases. When innovative analytics leaders have ideas, they often struggle to get internal support and momentum for the project because most people don’t really understand what AI can do. How do you demonstrate potential feasibility without actually building the solution? By generating data and rapid demos. This is the subject of Episode 18.

I’ve created a three step process for doing this quickly:

1. Hypothesize a problem.
2. Generate the data.
3. Build the demo.

This is actually quite easy with ChatGPT and custom GPTs, and I walk through an example in Episode 18. I've also compiled a guide with all these steps and a more complex RAG example available here.

‍