Blog

With the emergence of generative AI in our daily lives—and AI companies working on new devices that may soon sit alongside our smartphones, we're approaching a paradigm shift in how consumers interact with digital services. Explore some emerging protocols like MCP (Model Context Protocol) and A2A (Agent-to-Agent) that enable businesses to expose their services to personal agents, ensuring they remains relevant and accessible in the AI-first economy.

When using Large Language Models (LLMs) for enterprise needs, it's often desirable to have these models return structured outputs. This structured output facilitates parsing the generations for further processing or exposure through APIs. Currently, a common solution involves providing formatting instructions to the LLM in the System Prompt and employing regular expressions (regex) to parse the output. However, this approach can sometimes result in the LLM not returning an output that follows the provided instructions. Also, writing regex expressions might be challenging. A better technique has emerged to achieve this desired behavior, which we will explore in this article.

In the constantly evolving realm of Generative AI Chatbots, maintaining conversations safety and relevance presents an ongoing challenge. You may have experimented with employing System Prompt as a means to safeguard against off-topic or harmful conversations, only to encounter their limitations, particularly when using small LLMs. In this article, we'll delve into implementing Guardrails for both the inputs and outputs of a LangChain Chatbot.

Explore how to use SQLCoder-7B, a Large Language Model (LLM) that we will deploy on Amazon SageMaker, along with LangChain to perform Natural Language Querying (NLQ). We will see how to use LangChain to create a pipeline that prompts the LLM to generate an SQL query, retrieves data from a PostgreSQL database, and passes the results as a context to the LLM to obtain the final response.

Retrieval Augmentation Generation (RAG) can enhance the capabilities of a Large Language Model (LLM) by incorporating a domain-specific informations that provides additional knowledge when the LLM generates a response. This integration can eliminates the need to retrain the model for specific knowledge. In this article, we will explore the integration of RAG with a Llama 2 chatbot, deployed on Amazon SageMaker. We will leverage Amazon Bedrock to create embeddings, FAISS as vector store, and LangChain to establish the pipeline

Discover Tabby, an open-source AI coding assistant. The race to create AI coding assistants that help us write computer code faster and better is heating up, and one of the promising candidates in this exciting field is Tabby. In this article, we’ll delve into the process of deploying Tabby on AWS ECS using Terraform, then integrate it in VSCode, taking advantage of its capabilities to enhance our development workflow

Discover how to deploy your own private LLM chatbot using AWS SageMaker, Hugging Face, and Terraform. This article explores the growing world of open-source LLMs, with a focus on Llama 2, Meta's answer to ChatGPT. Learn how to set up the necessary infrastructure, configure SageMaker with Hugging Face, and seamlessly integrate LLM inference into your chatbot.

In this article, we'll explore the potential risks associated with utilizing Public Terraform modules and discuss ways to protect your AWS account from unauthorized access.