MLOps
MLOps is the discipline of operating machine-learning and AI systems reliably in production — covering data pipelines, model training, deployment, monitoring, drift detection, governance, and incident response — analogous to DevOps for traditional software.
Full definition
MLOps adds to DevOps the lifecycle concerns specific to AI: training data versioning, feature stores, model registries, A/B and shadow deployment, drift and bias monitoring, automated retraining, and reproducibility. Mature MLOps platforms include Vertex AI, SageMaker, Databricks, MLflow, Kubeflow, and Weights & Biases.
Why it matters
Most AI projects that fail in production fail because of MLOps gaps, not modeling gaps. Gartner has repeatedly reported that the majority of AI projects never reach production. Investing in MLOps capability is what separates organizations that ship AI from those that perpetually pilot.
Example
A retailer's pricing model is retrained nightly on the prior day's transactions, evaluated against a held-out test set, gradually rolled out to 5% then 25% then 100% of stores via a feature flag, and monitored for drift — all automated through an MLOps pipeline.
Related terms
- AI PipelineAn AI pipeline is the end-to-end sequence of stages that turns raw data into deployed AI predictions or content — ingestion, cleaning, feature engineering, training, evaluation, deployment, monitoring, and retraining — usually orchestrated as code.
- Model DriftModel drift is the degradation of an AI model's performance over time as the data it sees in production diverges from the distribution it was trained on — the silent failure mode of every deployed ML system.
- AI GovernanceAI governance is the framework of policies, roles, controls, and processes an organization uses to ensure its AI systems are lawful, safe, fair, accountable, and aligned with business intent — across the full lifecycle from problem framing to retirement.
- InferenceInference is the process of running a trained AI model on new, unseen inputs to produce predictions, classifications, or generated content — the part that runs every time a user interacts with the system.
- ModelAn AI model is the trained artifact — a specific set of numerical weights plus an architecture — produced when a learning algorithm runs against training data, and the thing that actually gets deployed and audited.
Source & further reading
Primary source: Sculley et al. — "Hidden Technical Debt in Machine Learning Systems" (NeurIPS) (2015).
Citation policy: this entry is part of the AIDOLS AI Implementation Glossary and may be quoted for research, journalism, and education with attribution to aidolsgroup.com/en/glossary/mlops/.