📄️ Guide for AI Newcomers
Large Language Models (LLMs) are powerful tools, representing a paradigm shift in how we build and use software. Machines now have the ability to reason and understand natural language as well as source code. We predict over the coming years, incumbents will either remake themselves, or be disrupted by AI-native products and platforms.
📄️ Rules of AI.JSX
AI.JSX uses the familiar JSX syntax, but it's not React.
📄️ Prompting in AI.JSX
Prerequisites
📄️ AI + UI
In traditional UI development, human engineers write deterministic code to handle every possible UI state. With "Just in Time" UI (JIT UI), human engineers produce building block components, then hand those to an AI to use in its response.
📄️ Observability
In this guide, we'll start with the Hello World example from the AI JSX template and iteratively add logging.
📄️ Performance
- Deployment Architectures - Discusses the various options (and the trade-offs) for how to build and deploy your AI.JSX powered applications.
📄️ Deployment Architectures
- Performance - Discusses trade-offs between performance and correctness. Presents five strategies for improving performance in your AI-powered apps.
📄️ DocsQA: Grounding Answers with a Source of Truth
Large Language Models (LLMs) are powerful tools for answering open domain questions, but they're prone to making things up ("hallucination") when their responses are unrestricted. In addition, publicly available LLMs are trained on public data and consequently won't have knowledge of your internal information (documents, data, documentation, etc.).
📄️ ESM
AI.JSX exports both CJS and ESM files. When you import or require ai-jsx, you'll get the file that matches the module system you're using. If you have a problem with this, please let us know by either dropping us a line in Discord or by opening an issue on GitHub.
📄️ JSX: Build System Considerations
Understanding these details is for power users. If you just want to get started quickly, clone the template repo and start experimenting.
📄️ MDX Output
By default, models emit text. If you ask them to, they'll emit markdown, which is a robust, easy way to provide more structured output to your user.
📄️ Models
AI.JSX supports models from OpenAI and Anthropic. Most people should start with OpenAI's GPT-4 model. If you need a very long context window, use Anthropic's Claude 100k.
📄️ Rendering
Once you assemble your AI.JSX component tree, you'll want to render it into text or UI components. The way you do it depends on how you're using AI.JSX.