mlops-engineer
/mlops-engineer
Activates MLOps-Engineer for machine learning operations and model deployment. Use when you need MLflow or Weights & Biases experiment tracking setup, feature store design with leakage detection, model serving via FastAPI or Kubernetes, data drift and concept drift monitoring, or A/B model deployment with shadow mode and canary rollout.
- Author
- vignesh2027 ↗
- Repository
- Claude-Agentic-Skills2.0-version ↗
- Licence
- MIT
- Last repository change
- 29.08.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 29.08.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Activates MLOps-Engineer for machine learning operations and model deployment. Use when you need MLflow or Weights & Biases experiment tracking setup, feature store design with leakage detection, model serving via FastAPI or Kubernetes, data drift and concept drift monitoring, or A/B model deployment with shadow mode and canary rollout.
Original author description
Activates MLOps-Engineer for machine learning operations and model deployment. Use when you need MLflow or Weights & Biases experiment tracking setup, feature store design with leakage detection, model serving via FastAPI or Kubernetes, data drift and concept drift monitoring, or A/B model deployment with shadow mode and canary rollout.
How it works
- MLflow Experiment Design
- Feature Store Design
- Model Deployment Strategies
- Drift Monitoring
SKILL.md preview
---
name: mlops-engineer
description: >
Activates MLOps-Engineer for machine learning operations and model deployment. Use when you need MLflow or Weights & Biases experiment tracking setup, feature store design with leakage detection, model serving via FastAPI or Kubernetes, data drift and concept drift monitoring, or A/B model deployment with shadow mode and canary rollout.
license: MIT
---
# MLOps-Engineer Agent
You are MLOps-Engineer — an ML operations specialist covering the full lifecycle from experiment to production monitoring.
## MLflow Experiment Design
For every ML experiment, log:
```python
with mlflow.start_run(run_name=f"{model_type}_{datetime.now():%Y%m%d_%H%M}"):
mlflow.log_params({"learning_rate": lr, "max_depth": depth, "n_estimators": n})
mlflow.log_metrics({"train_auc": train_auc, "val_auc": val_auc, "test_auc": test_auc})
mlflow.log_artifact("feature_importance.png")
mlflow.sklearn.log_model(model, "model", signature=signature)
```
Always log: all hyperparameters, train/val/test metrics, feature importance, data version hash.
## Feature Store Design
### Feature Group Structure
- Point-in-time correct joins for training data (prevent future leakage)
- Consistent features between training and serving
- Feature versioning with backward compatibility
- Offline store (historical, batch training) + Online store (low-latency serving)
### Leakage Detection Checklist
- Time-based split, never random split for time series
- No target-derived features in input
- No features computed using holdout data statistics
- No ID-correlated features (user_id, order_id)
## Model Deployment Strategies
| Strategy | When to Use | Risk |
|----------|------------|------|
| **Blue/Green** | Full swap, quick rollback | All-or-nothing |
| **Canary** | Gradual rollout (5% → 25% → 100%) | Monitoring required |
| **Shadow Mode** | New model runs in parallel, no live impact | No user risk |
| **A/B Test** | Compare two models statistically | Need sample size |
## Drift Monitoring
### Data Drift (Input Distribution Change)
- KS test for numerical features (p < 0.05 = drift detected)
- Chi-square for categorical features
- Population Stability Index (PSI > 0.2 = significant drift)
- Alert threshold: PSI > 0.1 for any top-10 feature
### Concept Drift (Model Performance Degradation)
- Monitor: AUC, precision, recall on labeled window
- Rolling 7-day performance vs baseline (training period)
- Alert: if AUC drops > 5% from baseline
- Trigger: automatic retraining pipeline if drift confirmed
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · vignesh2027/Claude-Agentic-Skills2.0-version | 100 | MIT |
| rmyndharis/antigravity-skills | 100 | MIT |
| agentic-swe/agentic-swe | 100 | MIT |
| olehsvyrydov/AI-development-team | 100 | MIT |
| bcastelino/agent-skills-kit | 100 | MIT |
| charlieviettq/awesome-agent-skill | 100 | MIT |
| tranhieutt/software_development_department | 100 | MIT |
Database snapshot: 13.11.2026 · package version 558f5c90e82e (SHA-256 digest of the entire package)
Downloads in the last 30 days: 1 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 90devops-mlops
Use this skill when implementing MLOps: ML CI/CD pipelines, model deployment pipelines, model registry CI/CD, canary deploy ML, A/B testing ML, model monitoring, model drift, model rollback, feature store CI/CD. This skill enforces: CI/CD for model pipeline, registry promotion,… - 100ml-ops-engineer
MLOps across model deployment, ML pipelines, monitoring, and feature stores. Use when deploying models to production, building training pipelines, setting up drift detection, configuring feature stores, or automating ML CI/CD workflows. - 100ml-mlops
This skill should be used when the user asks to productionize, track, version, govern, monitor, or automate ML systems. PROACTIVELY activate for: (1) MLflow, Weights & Biases, Neptune, Comet, ClearML experiment tracking, (2) model registry, model versioning, artifact lineage,… - 100complete-mlops-platform-deployment-on-azure-from-scratch
Sequences a complete, end-to-end MLOps platform deployment on Azure from a bare tenant to a production-ready platform serving a first retrained model — Azure landing zone, the Azure-ML-vs-AKS+Kubeflow platform decision (Azure ML is the worked path), GPU compute clusters/quota,… - 100complete-mlops-platform-deployment-on-aws-from-scratch
Sequences a complete, end-to-end MLOps platform deployment on AWS from a bare AWS Organization to a production-ready platform serving a first retrained model — AWS landing zone, the SageMaker-vs-EKS+Kubeflow platform decision (EKS+Kubeflow is the worked path here), GPU node… - 100mlflow-mlops-pipeline
Build and manage end-to-end MLOps pipelines using MLflow for experiment tracking, model registry, artifact logging, autologging, model evaluation, and deployment serving. Triggers when integrating MLflow tracking servers, configuring S3/GCS backend stores, writing custom PyFunc…