In this week's episode of AI That Works, we dive into the crucial art of user-defined prompting and how it can revolutionize your AI systems! We explore how to give your users the right level of control over prompts, leading to more flexible and effective AI applications.
Think of it like building a dashboard – you wouldn't hardcode everything, would you? We discuss how to apply that same principle to prompting, enabling users to define their own analytics and tailor the system to their specific needs.
We tackle the common tension: How do you balance user control with ensuring the system still works reliably? We share practical insights and examples, from schema generation to dynamic formatting, showing you how to bridge that gap.
Join us as we explore building AI pipelines that truly work for your users!
Check out our github: https://www.github.com/boundaryml/baml
AI That Works repo: https://github.com/ai-that-works/ai-that-works
Socials:
X: https://x.com/boundaryml
Discord: https://discord.com/invite/yzaTpQ3tdT
LinkedIn: https://www.linkedin.com/company/boundaryml/
⏱️ Chapters:
Intro
Getting a Good Answer
Form Builder Look
Terminology and Data Sanitization
Dynamic Structured Outputs
Patient Statuses Example
Zooming Out, Generalizing
Taking it Another Layer
Agentic Primer
Transcript
I've watched this one closely. Ask me anything about it and I'll point you to the exact moment.