A production-grade, open-source MLOps engine for real-time customer churn prediction. Features zero-downtime GitOps deployments via ArgoCD, Istio service mesh canary traffic splitting, MLflow tracking, Prometheus/Grafana drift monitoring, and automated retraining pipelines powered by Terraform IaC.
I built AegisML to solve the gap between training an ML model and running it reliably in enterprise production. Most templates stop at API deployment, but AegisML covers the entire lifecycle—automated drift detection, zero-downtime canary rollouts via Istio and ArgoCD, and infrastructure-as-code using Terraform. Excited to hear your thoughts and feedback!