Learning Resources
Governance is a clear record of who decided, on what evidence, and who answers for it. These modules teach it by doing: each one builds something small, measures it, and writes down the evidence.
How to tell whether a prompt is any good, and what actually improves it. Lessons for anyone who writes prompts, plus optional hands-on code with DSPy and a small AI model on your own computer.
The examples use open AI models that run on your own computer. No data leaves it, and each run costs nothing.
Every result on these pages comes from a run you can repeat yourself. That includes the ones that did not work.
Code checks the things that can be counted. People make the judgement calls, decide what is fair, and answer for the result.