AI & ML interests
AIOps, Kubernetes Reliability, Time-series Anomaly Detection, Root Cause Analysis, Explainable AI (XAI)
AIONYX: Artificial Intelligence for Resilient Ops
AIONYX (AI + Onyx) represents a fusion of Artificial Intelligence and the Onyx stoneโa symbol of strength, stability, and reliability. This project is an AI-driven AIOps platform designed to predict failures, detect anomalies, and optimize costs in Kubernetes-based systems.
๐ Vision
To build a production-grade MLOps platform that bridges the gap between infrastructure engineering (DevOps) and data-driven intelligence (Data Science).
๐ Tech Stack
- Orchestration: Dagster
- Storage: PostgreSQL (TimescaleDB potentially)
- Infrastructure: Kubernetes, Terraform
- CI/CD: GitHub Actions
- Language: Python
- ML: Scikit-learn, PyTorch (Anomaly Detection, Time-series Forecasting)
๐ Repository Structure
platform/: Core application logic (Dagster, ML pipelines, API).infra/: Infrastructure as Code (Terraform, K8s manifests).docs/: Thesis documentation, system architecture, and business roadmap.experiments/: Jupyter notebooks and datasets for research.
๐ Thesis Context
This repository serves as the primary codebase for the Master's Thesis: "An Intelligent AIOps Platform for Predictive Reliability and Cost Optimization in Kubernetes".
Developed as a Master's Thesis at Malmรถ University.