Upgrad

This practice project is all about learning by doing. Through a series of immersive, hands-on lab activities, you’ll build and deploy an intelligent assistant bot that runs on AWS’s modern serverless infrastructure. 

You’ll experiment with Retrieval-Augmented Generation (RAG) and Natural Language to SQL (NL-to-SQL) pipelines, gaining insight into how to blend unstructured knowledge retrieval and structured data querying within a single, user-friendly application. As you progress, you’ll configure cloud resources, create embeddings, build and test Lambda functions, and tie everything together with an interactive Streamlit app. 

By the end of the project, you’ll have designed and launched a fully functional AI-powered assistant that can answer questions from both documents and business data—arming you with the practical skills and confidence to implement similar solutions in real-world projects.

Azure Account with Restriction details. 

Resources are allowed only in these locations:

eastus, eastus2, Canada central and global

✅ Allowed Azure Services
1️⃣ Virtual Machines
Service:Microsoft.Compute/virtualMachines

Allowed VM SKUs:

Standard_B2ms

Standard_B1ms

Standard_B4ms

Standard_D2_v3

Standard_DS1_v2

2️⃣ App Service Plan
Service:in canadacentral onlyMicrosoft.Web/serverfarms

Allowed SKUs: F1,B1 and Standard

3️⃣ Web Apps
Service:in canadacentral only Microsoft.Web/sites

4️⃣ Storage Accounts
Service:Microsoft.Storage/storageAccounts

Allowed SKUs: Standard_LRS and StandardSSD_LRS

5️⃣ Azure Container Registry
Service:Microsoft.ContainerRegistry/registries

Allowed SKU:Basic

6️⃣ Azure Kubernetes Service (AKS)
Service:Microsoft.ContainerService/managedClusters

Allowed SKUs: Basic and Standard

7️⃣ Virtual Network
Service:Microsoft.Network/virtualNetworks

8️⃣ Application Insights
Service:Microsoft.Insights/components

9️⃣ Log Analytics Workspace
Service:Microsoft.OperationalInsights/workspaces

🔟 Resource Groups
Service:Microsoft.Resources/resourceGroups

Sandbox contains

AWS account access within Environment

Ubuntu VM with tools Docker Python, Vscode, Openai, Huggingface, autogen, awscli, openai, nfs-common, transformers, datasets, accelerate ,sentencepiece, pyautogen

Data Engineering with Databricks- Week 3

Data Engineering with Databricks- Week 2

Data Engineering with Databricks- Week 1

Playground Lab: Azure Data Engineering with Databricks 

Prompt Engineering

Prompt engineering is the process of designing and crafting prompts for large language models (LLMs) to get the desired output. LLMs are trained on massive datasets of text and code, and they can generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way. However, LLMs need to be prompted correctly in order to produce the best results.

How prompt engineering is helpful for content writers:

For content and blog writers, prompt engineering can be a game-changer. By using well-crafted prompts, writers can generate ideas, overcome writer's block, and even refine their writing style. It's a creative tool that can spark inspiration and assist in the content creation process.

• Brainstorming ideas: Prompt engineering can help content writers to brainstorm ideas for new articles, blog posts, and other types of content.
• Researching topics: Prompt engineering can help content writers to research topics more efficiently by providing them with summaries of key information.
• Generating content: Prompt engineering can help content writers to generate content more quickly and easily by providing them with templates and outlines.
• Improving the quality of content: Prompt engineering can help content writers to improve the quality of their content by ensuring that it is well-written, informative, and engaging.