For a business applications developer, one of the problems with working on solutions that use LLMs is that you often start with some kind of Python notebook or script where you work out whether something is possible. Once you’ve proved to yourself that it works, you then have the problem of how to show that to an audience of business users.
You might be able to get away with Python script for internal users, but it definitely won’t work for external users. Even internally many of the people you work with will need to see something wrapped up in a more user-friendly way.
This is where Superagent comes in. It’s an open-source framework for quickly building AI assistants and chatbots. Superagent adds a user interface and a bunch of other features that make it easy to build a chatbot that can be used by anyone. It also offers different options for hosting - either their cloud platform or ways to host locally, via replit or in your online hosting.
Superagent moves building prototypes from scripts into configuration - start a new chatbot, define some tools for it or give it some data sources to use, give it a system prompt to work with then get going trying it out. All online through an easy-to-use interface.
Superagent also allows agents to be defined programmatically, which might be more appropriate for more complex applications or for developers who are more comfortable with code.
For software developers Superagent also provides an API that you can use from your custom applications, letting Superagent handle the heavy lifting of LLM work and leaving you to just focus on the user interface.