Projects | Joe DevOps Lab

Architecture

AI Architecture

Architecture Highlights

React provides the web-based DevOps dashboard and AI Assistant interface.
FastAPI serves as the backend API and orchestration layer.
Ollama provides local LLM inference for AI-assisted DevOps operations.
Kubernetes integration enables cluster inspection and controlled operational actions.
PostgreSQL stores projects, users, and persistent DevOps action history.
Redis provides application-level caching and service support.
AI-assisted workflows interpret natural-language operational requests before executing approved actions.
Health endpoints provide visibility into application, database, Redis, and AI service availability.
Docker provides containerized application and service deployment.

Project Overview

🤖 AI-Assisted DevOps

This project demonstrates an AI-assisted DevOps platform designed to bring intelligent operational capabilities into a unified DevOps workflow. Instead of relying entirely on manual Kubernetes commands and log inspection, the platform provides a web interface where operational information can be inspected and DevOps actions can be initiated through natural-language requests.

⚙️ Platform Architecture

The platform combines a FastAPI backend, React frontend, Kubernetes integration, PostgreSQL persistence, Redis, and a locally hosted Ollama LLM. AI capabilities are focused on practical DevOps workflows including application log analysis and interpretation of natural-language operational requests.

☸️ Kubernetes Automation

The platform can inspect Kubernetes cluster resources such as nodes, pods, and deployments, while controlled DevOps actions can be executed through the platform and recorded in persistent action history. This provides a practical demonstration of how AI can be integrated with existing DevOps tooling without replacing established infrastructure and operational controls.

Technology Stack

FastAPI React Kubernetes PostgreSQL Redis Ollama Docker Python

Key Features

AI-Assisted Log Analysis
Natural-Language DevOps Action Interpretation
Kubernetes Cluster Inspection
Kubernetes Node, Pod & Deployment Visibility
Controlled Kubernetes Operational Actions
AI Action Preview Before Execution
Persistent DevOps Action History
React-Based DevOps Dashboard
FastAPI REST API
PostgreSQL Data Persistence
Redis Integration
Local LLM Inference with Ollama
Application & Service Health Monitoring
Containerized Application Architecture

Screenshots

AI DevOps Dashboard

AI Assistant

AI Analysis History

AI Action Intent

Key Challenges & Learnings

Integrating an LLM into a DevOps workflow required clear separation between AI interpretation and actual infrastructure operations.
Designing controlled AI-assisted actions helped ensure that natural-language requests could be reviewed before execution.
Kubernetes API integration provided practical experience in exposing cluster inspection and operational capabilities through an application layer.
Persisting DevOps actions created an auditable history of operational activity rather than treating AI interactions as temporary requests.
Running the LLM locally with Ollama demonstrated an approach for integrating AI capabilities without relying entirely on external AI APIs.
Building health checks across the application, database, Redis, and AI services improved operational visibility and reliability.
The project strengthened practical understanding of combining AI, APIs, Kubernetes automation, observability, and DevOps workflows into a single platform.

GitHub Repository ← Back to Projects

AI DevOps Platform