Deployment
The process of making a trained ML model available for use in production applications. Deployment involves packaging the model, setting up serving infrastructure, and establishing monitoring.
Why It Matters
Deployment is where models create value. An estimated 87% of ML models never reach production — bridging the gap from prototype to deployment is the biggest challenge in applied AI.
Example
Containerizing a model with Docker, deploying to Kubernetes, setting up auto-scaling, implementing health checks, and connecting monitoring dashboards.
Think of it like...
Like the difference between a prototype car in a lab and one rolling off the assembly line — deployment is the manufacturing process that turns experiments into products.
Related Terms
Model Serving
The infrastructure and process of deploying trained ML models to production where they can receive requests and return predictions in real time. It includes scaling, load balancing, and version management.
MLOps
Machine Learning Operations — the set of practices that combine ML, DevOps, and data engineering to deploy and maintain ML models in production reliably and efficiently.
API
Application Programming Interface — a set of rules and protocols that allow different software applications to communicate with each other. In AI, APIs let developers integrate AI capabilities into their applications.
Model Monitoring
The practice of continuously tracking an ML model's performance, predictions, and input data in production to detect degradation, drift, or anomalies after deployment.