AI/MLOps for Busy People: A Field Guide To Implementing Cloud Native AI/ML
Talkat Cloud Native University 2025 · on Tuesday, 2025-04-01· at 10:40· Level 0 | ICC Capital Hall | Room J·
summary
Maps ML lifecycle stages to concrete tools and practices: use Apache Airflow and dbt for repeatable ETL and DAG orchestration; validate and catalog with serabus and OpenMetadata; serve consistent features with Feast or Reddus as an online store and Milvvis as a vector database for embeddings; manage experiments, model artifacts and model registry with MLflow; automate packaging and retraining via GitHub Actions; deploy with KServe/KubeFlow and monitor drift with Prometheus and Evidently.
AI-written summary · about
chapters (10)
- 0:00MLOps introduction
- 2:08What is MLOps
- 7:22Experimental to production ML
- 10:48Data collection & preprocessing
- 13:39Feature engineering & validation
- 16:49Data cataloging & vector DBs
- 19:38Experimentation & model eval
- 23:09Model deployment & CI/CD
- 26:40Serving & inferencing
- 27:42Monitoring ML systems
related talks
13mMLOps: From AI Research to Production-Ready Systems
24mFrom Model to Production: Automating the ML Lifecycle with CI/CD
41mFrom Prototype to Production Building a Scalable MLOps Platform from a Reference Arch
26mKubernetes and MLOps for Scalable and Reproducible Generative AI
13mMLOps at Scale
21mMastering the DevOps Machine Learning Pipeline
18mAutomate the boring ML stuff with pipelines
47mMLOps tools to automate ML deployment