Posts

Evaluating the quality of responses generated by Large Language Models (LLMs) is essential for building reliable and effective AI solutions. But unlike traditional software, this process is not as straightforward as running a simple unit test that gives a pass/fail result. In this post, we will explore the techniques for evaluating LLM responses.

In my previous post using tools, I discussed how a Retrieval-Augmented Generation (RAG) system can be enhanced with plugins that handle specific queries using predefined functions. While this approach works well for a limited set of query types, it becomes difficult to scale as user queries become more diverse and complex. In this post, I’ll explore how we can use a plugin to generate dynamic SQL queries from natural language, enabling the AI assistant to answer a much broader range of questions.

In my previous post I discussed the concept of using a Retrieval-Augmented Generation (RAG) approach to create an AI assistant that can answer questions based on your own data. Retrieving the relevant records that has the same semantic meaning as the user query will not be sufficient for all use cases. In this post, I will discuss about the scenarios where we need more than just semantic search to answer the user queries. And one of the ways to solve those problems.

The serverless technology has been around for a while now, and it has revolutionized the way we build and deploy applications. Azure Functions, a serverless compute service from Microsoft Azure allows to run code on-demand without having to manage infrastructure. In this article, we will explore how running the Azure Functions on a dedicated Kubernetes cluster, such as Azure Kubernetes Services (AKS), can provide added benefits and flexibility to your applications.

A Virtual Assistant can assist users in answering questions, providing information and performing tasks. In the past, the virtual assistants were built using predefined rules and templates. This approach posed limitations to the number of tasks that the virtual assistant could perform and the quality of the responses it could generate.