No-code Decision Engine

for non-programmers

https://github.com/user-attachments/assets/c2b5353d-6dca-41dc-b0e3-8061d3cb4548

Summary

This decision engine can generate one final decision based on the user's decision tree. Next, this policy can be stored and executed in the back-end.

Prerequisites

For this project to run, make sure these software are installed:

• python

• nodejs

Installing

To run locally:

1. Clone this repo

   git@github.com:jefersoneiji/decision-engine.git
   

2. Install dependencies in your machine

   yarn install
   

3. Start both the front and back-end

   yarn dev
   

Testing

Execute the following command:

yarn test

Linting

yarn lint

Apps

• README | Frontend: Contains front-end logic for the policy editor
• README | Backend: Contains back-end logic for the CRUD operations in policies
• README | PolicyDB: Local database with sqlite and sqlalchemy
• README | ExecutionEngine: Responsible for processing and outputing one decision after executing a policy

Design Decisions

Turbo is used to reduce the number of steps required for executing commands such as spin up apps in dev mode.

Back-End

• For the sake of speed and simplicity. SQLite was choosen as local database engine
• Folders contain file according to their purpose. For instance routes only contains API routes. This structure make files predictible
• As a way to provent bugs and enforce code styling pytest was added to the project

Front-End

• Because of its reliability and simplicity Axios was chosen as http-client
• Most user feedback comes shows as Toast notifications from react-toastify
• As a way to provent bugs and enforce code styling eslint was added to the project
• An AI assistant may be helpful in answering user's questions related to docs and help them through problem solving while create policies

Made by Jeferson Eiji ➡️ Get in touch!