From 2010 to 2013 I was an instructor at the George Washington University, teaching a computer science capstone called “Senior Design” (CS 4244). The course was four months of taking an idea — something vague, full of unknowns — and turning it into a known, finished invention. Students came in knowing how to code. They left knowing how to build something real.
What took an entire semester — eight months of scoping, building, presenting, and iterating — can now be done in a day with AI. A student today can go from idea to working prototype before lunch. That’s the world we live in now.
And yet, these lessons stand the test of time. AI can write your code and generate your slides, but it can’t teach you how to pitch to someone who doesn’t care about your architecture. It can’t teach you perception, leverage, or when to push back on a client. The hard part was never the building — it was everything around it. That hasn’t changed.
In 2010 I put together a three-part series called “Idea to Invention” that became the backbone of the course. It covered everything from how to take a raw concept and scope it into something buildable, to the craft of presenting your work in a way that actually lands. Three years later, my own work had gotten so intense that I had to give up teaching. Before I left, I put together one final presentation — a 96-slide career talk that distilled everything I’d learned building a company from scratch.
The presentations
How to take a blank page and turn it into something real — scoping an idea, identifying your audience, and understanding why most technical people fail at the pitch, not the code.
Going deeper into the build — screencasting, prototyping, working with APIs, and the discipline of shipping something that isn’t perfect but is done.
The business side — pricing, client negotiation, and the uncomfortable reality that your client probably doesn’t care how clever your architecture is.
My farewell lecture — 96 slides covering the government job I left, the music school that needed “just a programmer,” a $70,000 accounting mistake, debugging a deployment from a Serbian internet cafe, and why your career has no structure unless you build it yourself.