Project Gemini: AI Development and Operations

This document provides a comprehensive overview of the Gemini project, its architecture, and development practices.

Project Overview

The Gemini project is a comprehensive AI development and operations platform. It consists of two main components:

AIML: A sophisticated AI/ML pipeline for training, deploying, and managing large-scale language models.
Open WebUI: A user-friendly web interface for interacting with the AI/ML models.

The project is highly organized and well-documented, with a clear separation of concerns between the different components.

Main Technologies

Frontend: SvelteKit, TypeScript, Tailwind CSS
Backend: Python, FastAPI, LangChain, Transformers
Database: PostgreSQL, Redis, MongoDB
Deployment: Docker, Kubernetes

Architecture

The project follows a microservices-based architecture, with the AIML and Open WebUI components running as separate services. The AIML component is further divided into a number of sub-components, each responsible for a specific part of the AI/ML pipeline.

Building and Running

Open WebUI

The Open WebUI component can be built and run using the following commands:

Install dependencies: npm install
Run in development mode: npm run dev
Build for production: npm run build
Run tests: npm run test:frontend

AIML

The AIML component is a collection of Python scripts and services. The elizabeth_cli.py script in aiml/02_models/elizabeth/deployment_configs/ provides an interactive CLI for the "Elizabeth" model.

Development Conventions

The project follows a number of development conventions, including:

Coding Style: The project uses prettier for code formatting and eslint for linting.
Testing: The project uses vitest for frontend testing and pytest for backend testing.
Contribution Guidelines: The project has a PULL_REQUEST_TEMPLATE.md file, which suggests that contributions are welcome.

Directory Overview

aiml/: The core of the project, containing the AI/ML models, training pipelines, and documentation.
open-webui/: The web interface for interacting with the AI/ML models.
platform/: A legacy or staging area for the aiml directory.
corpus-data/: Contains datasets for training the models.
deployment/: Contains deployment scripts and configurations.
docs/: Contains additional project documentation.
logs/: Contains logs from the various services.
novas/: Contains different versions of the "Nova" models.
open-webui-functions/: Contains serverless functions for the Open WebUI.
planner/: Contains planning documents for the project.
projects/: Contains related projects.
secrets/: Contains secrets and other sensitive information.
systemd/: Contains systemd service files for running the project's services.