Build an MLOps platform on top of OpenStack
summary
Model Arts runs on Open Telecom Cloud (OpenStack) and provides data labeling, versioning, AutoML (called XML), built-in algorithms and Docker/Jupyter support, plus GPU provisioning and OBS S3-style storage. A demo automation script uses OpenStack CLI to create a 100-image flower dataset, train a ResNet/TensorFlow model, and deploy a real-time service — predictions hit 100% for a tulip and 97% for a daisy. Claims include two‑minute deployment, AIC4 compliance work, and an OpenStack‑driven automation pipeline.
AI-written summary · about
chapters (6)
- 0:00Introduction to the team
- 1:20Open Telecom Cloud overview
- 4:28Model Arts platform features
- 9:50Medical AI demo
- 14:40AI pipeline automation demo
- 25:12Q&A and session conclusion
related talks
36mMLOps from the edge to on demand visualization
29mTeam Teardown: MLOps in the Cloud for IoT
53mExperience with an end-to-end MLOps pipeline: Practical insights and 3D architectural perspectives
36mDataOps and MLOps: NLP Development and Deployment in Kubernetes made easy
35mBuilding on-Premises MLOps for ISS Columbus Ground Operations - Samo Turk
50mMLOps for a hybrid AI platform compliant with European law
41mMLOps in 30 Minutes
13mMLOps: From AI Research to Production-Ready Systems