Small fixes in lessons 1, 2, and 4

Author: PabloNunes

01-IntroToGenAI/images/semantic-kernel-full-loop.png

@@ -1,6 +1,6 @@
-# Getting Started with AI Development Tools
+# Lesson 1: Getting Started with AI Development Tools
 
-Refresh your generative AI knowledge and understand the .NET tooling available to help you to develop generative AI applications.
+*Refresh your generative AI knowledge and understand the .NET tooling available to help you to develop generative AI applications.*
 
 ---
 
@@ -18,7 +18,7 @@ Refresh your generative AI knowledge and understand the .NET tooling available t
 
 ---
 
-## Lesson 1 - Generative AI Fundamentals for .NET
+## Generative AI Fundamentals for .NET
 
 Before we dive in to some code, let's take a  minute to review some generative AI (GenAI) concepts. In this lesson, **Generative AI Fundamentals for .NET**, we'll refresh some fundamental GenAI concepts so you can understand why certain things are done like they are. And we'll introduce the tooling and SDKs you'll use to build apps, like **MEAI** (Microsoft.Extensions.AI), **Semantic Kernel**, and the **AI Toolkit Extension for VS Code**.
 
@@ -159,4 +159,4 @@ In the next chapters, we'll explore these scenarios in detail, providing hands-o
 
 Next up, we'll get your development environment setup! So you'll be ready to dive into the world of generative AI with .NET!
 
-👉 [Set up your AI development environment](/02-SettingUp.NETDev/README.md)
+[Set up your AI development environment](/02-SettingUp.NETDev/README.md)

01-IntroToGenAI/readme.md

@@ -1,8 +1,8 @@
-# Setting Up the Development Environment for This Course
+# Lesson 2: Setting Up the Development Environment for This Course
 
-This lesson will guide you through setting up your development environment for this course. To ensure your success we've prepared a devcontainer configuration that will provide all the tooling you need to complete the course. You can run the devcontainer in GitHub Codespaces (recommended) or locally on your machine. And we also demonstrate how to set up your GitHub access tokens to interact with GitHub Models.
+*This lesson will guide you through setting up your development environment for this course. To ensure your success we've prepared a devcontainer configuration that will provide all the tooling you need to complete the course. You can run the devcontainer in GitHub Codespaces (recommended) or locally on your machine. And we also demonstrate how to set up your GitHub access tokens to interact with GitHub Models.*
 
-If you rather not use GitHub Models to interact with LLMs, we have you covered there too with guides to setup Azure OpenAI and Ollama.
+We have you covered with guides to setup Azure OpenAI and Ollama, if desired.
 
 ---
 
@@ -38,9 +38,9 @@ Here's a quick rundown of the services:
 - **Azure OpenAI**: A paid service that provides access to a wide range of AI models. It includes all of the benefits that you'd come to expect from Azure include robust security and scalability. This is a great option for this course if you already have access to an Azure OpenAI subscription.
 - **Ollama**: Ollama allows you to run AI models locally on your machine or within a Codespace or devcontainer for free. It's a great option if you prefer to run the models locally, but it is slower than the cloud-based options.
 
-👉 GitHub Models right for you? Follow the rest of this document.
-👉 Azure OpenAI have your eye? [This is the document for you](getting-started-azure-openai.md).
-👉 Ollama your choice? [This guide has the info you need](getting-started-ollama.md).
+> If **GitHub Models** is your choice, follow the rest of this document to set up your development environment with GitHub Models.
+> - **Azure OpenAI** have your eye? [This is the document for you](getting-started-azure-openai.md).
+> - **Ollama** your choice? [This guide has the info you need](getting-started-ollama.md).
 
 ## Learn and test AI models with GitHub Models
 
@@ -141,11 +141,14 @@ In this lesson, you learned how to set up your development environment for the r
 
 ### Additional Resources
 
+- Test this guide with other LLMs orquestrators!
+    - [Azure OpenAI](getting-started-azure-openai.md)
+    - [Ollama](getting-started-ollama.md)
 - [GitHub Codespaces Documentation](https://docs.github.com/en/codespaces)
 - [GitHub Models Documentation](https://docs.github.com/en/github-models/prototyping-with-ai-models)
 
 ## Next Steps
 
 Next, we'll explore how to create your first AI application! 🚀
 
-👉 [Core Generative AI Techniques](../03-CoreGenerativeAITechniques/readme.md)
+[Core Generative AI Techniques](../03-CoreGenerativeAITechniques/readme.md)

02-SettingUp.NETDev/readme.md

@@ -15,7 +15,10 @@
 
 **Index**
 
-1. [eShopLite Demos](#intro-video)
+1. [Intro Video](#intro-video)
+1. [eShopLite Demos](#eshoplite-demos)
+    - [eShopLite with Semantic Search](#eshoplite-with-semantic-search)
+    - [eShopLite with Realtime Analysis](#eshoplite-with-realtime-analysis)
 1. [Chat with your Data](#understanding-github-codespaces)
 1. [Creative Writer](#pre-flight-check-setting-up-github-access-tokens)
 1. [Conclusions and Resources](#conclusions-and-resources)
@@ -40,8 +43,6 @@ The first demo, we show how to use the Semantic Kernel to enhance the search cap
 
 ### eShopLite with Semantic Search
 
-![Video thumbnail for eShopLite with Semantic Search](./images/eshoplite-semantic-search.png)
-
 In eShopLite with Semantic Search, we use the Semantic Kernel to enhance the search capabilities of the e-commerce application. Semantic Kernel auxiliate us to create a more robust search engine that can understand the context of the user's queries and provide more accurate results. For example, if a user searches for "do you have something for cooking", the search engine can understand that the user is looking for kitchenware and show the most relevant products, in context of our sample, it returns Camping Cookware. 
 
 ![Image demonstrating the search capabilities in eShopLite](./images/search-eshoplite.png)
@@ -202,7 +203,7 @@ try
 }
 ```
 
-With the code above, we  generate the embedding for the search query, search the vector database for the most similar product, and get a response message using the found product information. Help the user find the products they need more easily, leading to a better shopping experience and increased sales. Moreover, as generative AI evolves, we need some telemetry and monitoring to understand the user's behavior and improve the search engine, this is where Azure Application Insights and .NET Aspire come in.
+With the code above, we generate the embedding for the search query, search the vector database for the most similar product, and get a response message using the found product information. Helping the user find the products they need more easily, leading to a better shopping experience and increased sales. Moreover, as generative AI evolves, we need some telemetry and monitoring to understand the user's behavior and improve the search engine, this is where Azure Application Insights and .NET Aspire come in.
 
 ![Image demonstrating the .NET Aspire tracing capabilities](./images/aspire-tracing-eshoplite.png)
 
@@ -216,8 +217,6 @@ Application Insights provides a comprehensive set of telemetry data, helping us
 
 ### eShopLite with Realtime Analysis
 
-![Video thumbnail for eShopLite with Realtime Analysis](./images/eshoplite-realtime.png)
-
 In eShopLite with Realtime Analysis, we use the Realtime audio capabilities of GPT-4o to analyze the conversations between the customer and the chatbot, providing a more personalized and engaging experience. For example, if a customer asks for a product recommendation, the chatbot can analyze the customer's request in real-time and provide a more accurate and relevant response.
 
 ![Image demonstrating the Realtime Analysis in eShopLite](./images/realtime-analysis-eshoplite.gif)
@@ -286,9 +285,6 @@ See a local demo of the feature as File Search in action:
 
 ## Chat with your Data
 
-![Video thumbnail for Chat with your Data](./images/chat-with-your-data-video.png)
-
-
 Do you want to chat with your data? In this demo, we'll use the Chat with your Data application to generate a conversation with the user's data, using a simple interface to upload a file and extract insights from it. 
 
 ![Image demonstrating the Chat with your Data application](./images/chat-with-your-data.png)
@@ -302,13 +298,18 @@ Do you want to chat with your data? In this demo, we'll use the Chat with your D
 
 ## Creative Writer Agent
 
-![Video thumbnail for Creative Writer Agent](./images/creative-writer-agent-video.png)
-
 Agents are a big topic in the current AI landscape, and to demonstrate their capabilities, we'll use the Creative Writer Agent, a tool that can generate creative and engaging text based on the user's input, helping to write researched, specific, and engaging content.
 
 ![Image demonstrating the Creative Writer Agent](./images/creative-writer-agent.png)
 
-This AI application is composed of four main components: Researcher, Marketing, Writer, and Editor. The Researcher component is responsible for understanding context, gathering information with a Bing search, and then summarizing the information. The Marketing component is responsible for understanding the user's intent, formulate queries, and retrieve information from the Vector DB. The Writer component is responsible for gathering the information from the other components and produce the article, while the Editor component is responsible for reviewing the article, providing feedback, it decides if the article is ready to be published or not. All of this behavior is orchestrated by Semantic Kernel, Microsoft AI Extension, and .NET Aspire.
+This solution centers on four dedicated modules that combine to generate high-quality content:
+
+- Researcher: Leverages Bing search to gather context, topics, and data, then concisely summarizes it.
+- Marketing: Interprets user intent, constructs relevant questions, and taps into the Vector DB for precise results.
+- Writer: Synthesizes findings from Researcher and Marketing, producing a cohesive writing of the article.
+- Editor: Assesses the draft, offers corrections, and decides whether it’s publication-ready.
+
+The workflow integrates relevant data, effective messaging, and review, being orchestrated by Semantic Kernel, Microsoft AI Extension, and .NET Aspire.
 
 ![Image demonstrating the Creative Writer Agent architecture](./images/creative-writer-agent-architecture.png)