Agent skill · software engineering · aj-geddes
Model Deployment
Deploy machine learning models to production using Flask, FastAPI, Docker, cloud platforms (AWS, GCP, Azure), and model serving frameworks
Why this skill is useful
Adds executable scripts for deploying machine learning models with Flask, FastAPI, and Docker, which are not commonly found in public documentation.
What it needs
Requires docker, fastapi, joblib, mlflow, scikit-learn, uvicorn installed locally. About 4k tokens when loaded. Last updated 2026-03-04. 307 stars on the source repository.
What this skill does
Model Deployment Overview Model deployment is the process of taking a trained machine learning model and making it available for production use through APIs, web services, or batch processing systems. When to Use When productionizing trained models for real-world inference and predictions When building REST APIs or web services for model serving When scaling predictions to serve multiple users or applications When deploying models to cloud platforms, edge devices, or containers When implementing CI/CD pipelines for ML model updates When creating batch processing systems for large-scale predictions Deployment Approaches REST APIs: Flask, FastAPI for synchronous inference Batch Processing: Scheduled jobs for large-scale predictions Real-time Streaming: Kafka, Spark Streaming for continuous data Serverless: AWS Lambda, Google Cloud Functions Edge Deployment: TensorFlow Lite, ONNX for edge devices Model Serving: TensorFlow Serving, Seldon Core, BentoML Key Considerations Model Format: Pickle, SavedModel, ONNX, PMML Scalability: Load balancing, auto-scaling Latency: Response time requirements Monitoring: Model drift, performance metrics Versioning: Multiple model versions in production Python Implementation Deployment Checklist Model format and serialization Input/output validation Error handling and logging Authentication and security Rate limiting and throttling Health check endpoints Monitoring and alerting Version management Rollback procedures Cloud Deployment Options AWS: SageMaker, Lambda, EC2 GCP: Vertex AI, Cloud Run, App Engine Azure: Machine Learning, App Service Kubernetes: Self-managed on-premises Performance Optimization Model quantization for smaller size Caching predictions Batch processing GPU acceleration Request pooling Deliverables Deployed model endpoint API documentation Docker configuration Monitoring dashboard Deployment guide Performance benchmarks Scaling recommendations
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aj-geddes/model-deployment