nao is a framework designed for building and deploying analytics agents. It allows users to create a custom context for their agent using the nao-core CLI, where they can include data, metadata, modeling elements, and rules. Users can also deploy a chat interface that enables anyone to interact with the agent and perform analytics tasks easily. For data teams, nao offers several features. The Open Context Builder lets users create a flexible context that can hold various components, such as data and tools, without restrictions. The platform is compatible with any data stack, ensuring versatility across different environments. Teams can also test their agents before deployment to ensure reliability and track performance over time, incorporating user feedback for continuous improvement. The self-hosted nature of nao ensures that data security is a priority, allowing users to manage their own LLM keys. Business users benefit from the framework's natural language processing capabilities, enabling them to ask questions in plain English and receive immediate analytics responses. The chat interface supports data visualization and provides transparency in how the agent derives insights. Users can easily give feedback on the accuracy of the agent’s responses, fostering a collaborative environment between business and data teams. Setting up nao involves several steps, starting with installing the nao-core package and initializing a project. Users can customize their setup by connecting databases, adding repositories, and configuring LLM keys. The process includes synchronizing context files and launching the chat interface. For those using Docker, there are specific commands for running the nao chat either with a bundled example project or a local project. The article also references additional resources for deployment and development, emphasizing community engagement through contributions and feedback.