Welcome to part two of our series about using Temporal to improve the reliability of applications built around Large Language Models (LLM) like the one that powers ChatGPT. Part one explained how to build a Temporal Workflow to process a series of documents and make them accessible to your LLM. This post will show how to develop a Temporal Workflow to find documents relevant to a user’s query and supply them as context to a prompt sent to the LLM using Context Injection. You’ll also learn how Temporal's abstraction will make your application more reliable and make it easier for you to extend it with new features.

Taking advantage of the burgeoning AI trend, many of today's applications are built around AI tools like ChatGPT and other Large Language Models (LLMs). AI-optimized applications often have complex software pipelines for collecting and processing data with the LLM. Temporal provides an abstraction that can significantly simplify data pipelines, making them more reliable and accessible to develop. In this post, you’ll discover why you should use Temporal to build applications around LLMs.