Applied AI with Eclipse Graphene
summary
Eclipse Graphene packages reusable AI modules as Docker images with protobuf/gRPC interfaces so modules are language neutral and stubs generate automatically. A web portal manages catalog and design studio while execution runs in separate environments including a public AI playground with one‑click deployment and a generic orchestrator that dispatches messages and runs containers it has never seen. Features include supports cyclic graphs for feedback loops, Jupyter Connect for interactive exploration, and initial reproducibility and benchmarking metadata across a marketplace of over 300 models and pipelines.
AI-written summary · about
chapters (15)
- 0:00Eclipse Graphene intro
- 1:33Portal vs execution environments
- 3:39AI pipelines for all
- 5:52Protuff & gRPC for containers
- 7:16Parallel pipeline orchestrator
- 9:47Data broker in pipelines
- 11:55Cyclic topologies in AI
- 13:43Jupyter Connect for models
- 15:34Benchmarking Graphene
- 17:23LLM pipelines & enhancements
- 19:14Graphene use cases
- 21:45Graphene funding & future
- 22:50Open source & community
- 23:10Live demo: Graphene AI4
- 28:50Q&A
related talks
28mBuilding AI Pipelines with Eclipse Graphene
30mDeveloping AI Integration in Domain Specific Tools with Theia AI
38mIntroducing Open Platform for Enterprise AI
29mOpen Source Sovereign AI: Building Enterprise GenAI Platforms for Control & Comp...
33mOperationalizing GenAI Applications with DevOps Practices
20mArchitecting for Innovation: Building AI Powered Apps and Platforms
38mIntroducing Open Platform for Enterprise AI
01h 03mBuild & deploy AI-powered apps
