AI

The Semantic Kernel SDK is an open-source SDK that allows developers to integrate large language models (also known as LLMs) in their applications. The Semantic Kernel SDK allows developers to integrate prompts to LLMs and results in their own applications. For example, say we’re developing a booking system for a medical clinic. Instead of creating a LLM from scratch, we can use the Semantic Kernel to use existing LLMs and create an AI agent that can understanding the natural language queries that our patients make, provide recommendations based on their queries, and book them in to appointments. ...

Azure OpenAI provides developers with the ability to add AI to their applications using a variety of different models from OpenAI. This includes GPT-4, GPT-4 Turbo with Vision, GPT-3.5-Turbo and Embedding models. We can add AI functionality to our applications using C#, Python or REST APIs. The Generative AI capabilities that are available in Azure OpenAI are provided through the models, which belong to different families. This article assumes that you already have access to Azure OpenAI. To use Azure OpenAI, you need to be approved. Luckily for me, my work already has a resource for me to use. If you haven’t got one, check out this guide to get started. ...

In Azure AI Search, semantic ranking improves our searches by using language understanding to rerank search results. Semantic search is a collection of query capabilities that improve the quality of search results using text-based queries. Using semantic search we can: Improve Search Results by adding a ranking over initial search results using advanced algorithms that consider the context and meaning of the query, resulting in more relevant search outcomes. Provides Additional Information by extracting and displaying captions and answers from search results, which can be used to improve the user’s search experience. In this article, we’ll talk a little about Semantic Search is in Azure AI Search, its advantages and limitations, how to set it up in Azure AI search, and how we can perform semantic searches using C#. ...

How much text data do you produce on a daily basis? This could be in the form of an email you send to a colleague, or posts you send on social media. You may have gone to a new restaurant and given them a review online, or you could have created some documents such as contracts as part of your work. Now imagine all that data being created on a global scale. How can we apply AI to this vast amount of text data to extract insights from it? ...

Azure Cognitive Search is a search service in Azure that gives you as a developer the tools for building search experiences over private data in your enterprise, web and mobile applications. Common use cases for search can include any scenario that surfaces text to users, such as searching a product catalog, or searching documents within your organization. Azure Search provides you with a bunch of cool capabilities, including: Full text and vector search over search indexes containing your content. Indexing, with optional AI enhancement for extracting and transforming your content. Query syntax for text search, autocomplete, geo-search, vector queries and more. Access to search via SDKs and REST APIs Integration with Azure AI services. In this article, I’ll cover what Azure Cognitive Search is, what components make up Azure Search, and where it can fit within your architecture. We’ll then create a Cognitive Search service using Bicep along with a storage account that we will pull our data from into the index. Once that’s deployed, we’ll start to work with our Cognitive Search service using C#, creating our data source for our indexer, creating the index and then querying our index with some simple queries. ...