Competencies: AI & ML > MLOps

MLOps Fundamentals

ML lifecycle, reproducibility, CI/CD for ML, model governance, team collaboration.

Critical

MLOps Engineer, ML Engineer

Experiment Tracking

MLflow, Weights & Biases, Neptune, metrics logging, artifact management, comparison.

Critical

MLOps Engineer, Data Scientist

Model Registry

Model versioning, staging, promotion, metadata, lineage, MLflow Model Registry.

High

MLOps Engineer, ML Engineer

Feature Stores

Feature engineering at scale, Feast, Tecton, online/offline stores, feature pipelines.

High

MLOps Engineer, ML Engineer

Model Serving

REST APIs, FastAPI, TorchServe, TensorFlow Serving, latency optimization.

Critical

MLOps Engineer, ML Engineer

Model Monitoring

Data drift, concept drift, performance degradation, alerting, retraining triggers.

Critical

MLOps Engineer, Data Scientist

Data Version Control

DVC, data pipelines, data lineage, dataset versioning, experiment reproducibility.

High

MLOps Engineer, Data Engineer

ML Pipelines

Kubeflow, Airflow for ML, Prefect, pipeline orchestration, DAGs.

High

MLOps Engineer, Data Engineer

GPU Infrastructure

CUDA, GPU clusters, cloud GPUs (A100, H100), multi-GPU training, cost optimization.

High

MLOps Engineer, ML Engineer

Model Optimization

Quantization, pruning, distillation, ONNX, TensorRT, inference optimization.

High

MLOps Engineer, ML Engineer

No personal status recorded

 — 

 — 

 — 

 —