Overview
This course is about using AI to do the work of a finance professional. It is intended for students who want a career in the financial sector and expect AI to be part of the job from the first day.
We will do three things. First, we will build basic AI literacy, to the extent that it is useful for financial work. Second, we will practice applying AI to the core tasks of finance. Third, we will build: a final project in groups that answers a real question in finance with AI, from raw data to a finished result.
Why take this course?
Employers now use these tools in their daily work, and they expect new hires to arrive with practical experience rather than general familiarity. They need someone who can produce financial analysis with AI, who knows where the tools fail on financial data, and who takes responsibility for the result. That expectation falls most directly on the newest members of a team, and this course is designed to help you meet it.
At the same time, I hope this course helps you build a “brand” of your own. As AI takes over more of the analysis, the analysis itself will do less to set you apart as the people you compete with will have the same tools. What sets you apart will inevitably have to be harder to copy: a story about what you have done and why, a point of view about where the tools help and where they mislead, and a reputation as someone colleagues want on their team.
One idea that has come out of the AI industry is the “builder mindset,” the view that a person with an idea and these tools can now make things that used to require a team. Throughout the course, I will ask you to think about what it would mean to build something in the finance role you are heading toward. The final project gives you and your group a place to develop one idea, so that each of you leaves with something concrete you built and can talk about.
It is also useful to say what this course is not. It is not a course on any particular AI product. The tools change every few months, so we will use whichever frontier assistant is best when the term begins, and the skills you develop should carry over to whatever replaces it. It is not a course on how AI models are built or on how a firm should deploy AI across an organization. Columbia already offers courses on both. It is not a programming course. No coding background is required, and you will not write code, though you will read and run code that AI writes for you.