Duque Ortega Mutis
Community@duqueom · Mexico City, Mexico
MLOps & Production ML | CI/CD, Docker, Kubernetes, Terraform, AWS | Turning models into reliable systems
Agent Skills by Duque Ortega Mutis
Showing 16 vetted skills indexed across 1 GitHub repositories.
drift-detection
Detect data and concept drift in ML services using PSI and sliced AUC metrics.
performance-degradation-rca
Merge metrics, drift, deployment history, and data provenance to document root causes of production performance degradations.
deploy-aws
Automate ML service deployments to AWS EKS with Kustomize overlays and IRSA bindings.
cost-audit
Analyze monthly cloud spend to detect budget overruns and recommend optimizations.
deploy-gke
Automates deployment of ML services to Kubernetes using GCP's GKE with managed identity.
rule-audit
Identify AGENTS.md invariant D-01 through D-27 deviations with file:line evidence.
batch-inference
Automate batch inference scoring millions of rows into partitioned Parquet files.
release-checklist
Automate release governance and pre-deployment checks for GCP and AWS pipelines.
model-retrain
Automate model retraining workflows with quality gates and MLflow promotion.
eda-analysis
Run a six-phase EDA pipeline on raw datasets to generate training, schema, and drift artifacts.
security-audit
Detect security flaws in code, configurations, and containers before deployment.
secret-breach-response
Detect leaked credentials in repositories, logs, or artifacts and initiate incident response.
rollback
Revert production ML deployments during incidents using Argo Rollouts or Kubernetes.
new-service
Scaffold a production-ready ML service with CI/CD and infrastructure templates.
concept-drift-analysis
Correlate sliced AUC/F1 metrics with ground-truth labels to diagnose concept drift.
debug-ml-inference
Diagnose latency and incorrect predictions in production ML inference services.