It had to happen. Adding an AI agent to help build rules has been on my list for a long time but I finally jumped in and worked on it.
Rules usually come from Natural Language
The initial requirement for a rule typically comes in a rambling email with poorly defined facts and incomplete conditionals. ifServ’s AI Assistant (agent) lets you describe your rule in detail with all the ifs, elses and whatevers. You can tell it what to expect from your input fact model and tell it what sort of data you would like in its response. The Agent builds a rule and a Test Suite for the rule that tests various scenarios of input facts.
If you don’t want to start creating an AI prompt from scratch, I’m creating some Template Prompts to create rules in various domains like, finance, travel, logistic, etc. to get you started.
Show Me
Here are a couple test runs of the AI Agent. It’s two typical use-cases for rules:
- A Customer Discount Computation for an Order based on various factors including order amount, seasonal discounts, etc.
- Travel Approval based Flight Time, Traveler status and other factors.
Order Discount
Travel Approval
Try It
The ifServ app is running here. It is still very much a beta version and I make no promises on stability or uptime. But I am continuously working on it. You can create a free account there and even invite other users to your “org”. There are transaction limits while the app is in this development stage. It is for evaluation and bug reporting only.