Practical End-to-End AI Development using Prompty and AI Studio
summary
Shows a concrete workflow: a Prompty spec (markdown + front matter) stored in VS Code with a VS Code extension, executed via Prompt Flow or LangChain/Semantic Kernel, authenticated with AAD. Demonstrates building a three‑agent app (research, product, writer) that generates a publishable camping blog post in seconds, plus an evaluator prompty that scores relevance (example scores flip between 3 and 2), proving debugging, tracing, and iterative prompt engineering truly work end to end.
AI-written summary · about
chapters (11)
- 0:00Practical AI development intro
- 4:51LLMs as language calculators
- 6:16What Prompty is
- 9:09Prompty in orchestration
- 11:41Getting started in VS Code
- 19:11Customizing prompts
- 21:19Prompty with LangChain/Semantic Kernel
- 24:17Custom evaluator prompts
- 32:30Prompty as a function
- 35:37Multi-agent blog app
- 42:23Summary and next steps
related talks
01h 08mIn Prompts We Trust
29mSupercharge Your Productivity: Mastering Prompt Engineering
42mCode First LLMOps With Azure AI & Prompty
27mEnhancing AI-Driven Quality Assurance
45mTest Driven Prompt Programming: How AI is Changing the Way We Code
47mThe Art of Prompt Engineering: Best Practices for Effective AI Interactions
44mSupercharge Your Productivity Mastering Prompt Engineering
31mPrompt Like a Pro & Vibe Code with AI Tools