RenAI

🌟 Overview

RenAI is a platform aiming to automatize development of software projects. It centers around an LLM agent that iteratively builds a codebase according to user specificatios, interacting with a dedicated environment and outside sources.

✨ Key Features

• Full Stack Web Application Building: The LLM agent develops both backend and frontend projects in a variety of modern frameworks (eg. Spring Boot, Angular).
• App Specification: Users can provide detailed specifications for their desired application, including technologies, coding styles and code templates, or rely on the agent to make choices.
• Agent Lifecycle Management: RenAI provides an intuitive interface for managing the agent's workflow and interacting with its environment.
• Configurable Infrastructure: Users can run the agent on common infrastructure to minimize costs or set up an isolated environment for security.
• Testing: The agent regularly writes tests and runs them to ensure code quality.

🚀 Quick Start

RenAI is not yet in production, but you can run the code locally following these steps:

1. Ensure you have Docker, Minikube, KubeCTL, Helm and Java JDK 21. You also need a newer version of Python and at least 6GB of free RAM for the LLM model inference.
2. Fork and fetch the RenAI backend repository and open a terminal in its root.
3. You need to build a Docker image for each of the core microservices: renai-core, renai-developer, renai-gateway and renai-llm-inference. For instance, for renai-core, you need to:

• navigate to the microservice root: cd services/renai-core
• Build the Java project: mvn clean package -DskipTests
• Build the image: eval $(minikube docker-env) and docker build -t renai-core:latest . The renai-llm-inference service is a Python project so you can skip the second step.

4. Start Minikube and install Kubernetes deployments with Helm: helm install std-release .. Scale each service as needed by modifying the replicaCount in kubernetes/dev/helm/values.yaml.
5. To interact with the system, port-forward the API gateway, for instance kubectl port-forward service/std-release-renai-gateway 8080:8080. You can now make calls to localhost:8080
6. Fetch the Web Frontend repository and run ng serve. You can now access the app at localhost:4200 and run your RenAI developer!

Status

In mid stages of development.

Contributing

All contributions are warmly welcomed. Head over to CONTRIBUTING.md for details.