KCD Colombia - Cloud Native Development in the AI Era

Transcript

1. kevindubois Cloud Native Development in the AI Era Kevin Dubois (@kevindubois) Senior Principal Developer Advocate, IBM
2. kevindubois Kevin Dubois ★ Sr. Principal Developer Advocate at ★ Technical Lead, CNCF DevEx TAG ★ Java Champion ★ From Belgium / Live in Switzerland ★ English, Dutch, French, Italian youtube.com/@thekevindubois linkedin.com/in/kevindubois github.com/kdubois @kevindubois.com
3. kevindubois
4. kevindubois
5. kevindubois Average developer trying to download, run, experiment & manage models, configure serving runtimes,ensure correct prompt templates, integrate it in their code… (Colorized, 2025)
6. kevindubois 8 Find LLMs Try prompts Experiment with your data Benchmarking Ideation & Prototyping Find the right stack Connect with Model & Prompt Retrieval Augmented Generation (RAG) Guardrails Agents, MCP, A2A, … Building and Integrating Chaining & Memory ... Model serving Endpoints Monitoring Operationalizing App Deployment Platform Incorporation The Journey of Developing with Generative/Agentic AI
7. kevindubois 9 Find LLMs Try prompts Experiment with your data Benchmarking Ideation & Prototyping The Journey of Developing with Generative/Agentic AI
8. kevindubois
9. kevindubois
10. kevindubois We’ve been here before…
11. kevindubois Operating Systems & App Servers/Languages cloud computing and automation artificial intelligence and machine learning
12. kevindubois
13. kevindubois Find models on Hugging Face
14. kevindubois
15. kevindubois While OpenAI open sourced AI models in its early days, the company has generally favored a proprietary, closed source development approach. The latter strategy has helped OpenAI build a large business selling access to its AI models via an API to enterprises and developers. However, CEO Sam Altman said in January he believes OpenAI has been “on the wrong side of history” when it comes to open sourcing its technologies. “
16. kevindubois Using, experimenting, and benchmarking these “Open Source” models
17. kevindubois Podman AI Lab Jlama
18. kevindubois 23 Find LLMs Try prompts Experiment with your data Find the right stack Connect with Model & Prompt Retrieval Augmented Generation (RAG) Guardrails Benchmarking Agents, MCP, A2A, … Building and Integrating Ideation & Prototyping Chaining & Memory ... The Journey of Developing with Generative/Agentic AI
19. kevindubois A simplified landscape Left / Right of the Model
20. kevindubois Model Context Protocol (MCP) LangChain4j MCP Client
21. kevindubois 26 Model serving Endpoints Monitoring Operationalizing Find the right stack Connect with Model & Prompt Retrieval Augmented Generation (RAG) Guardrails Agents, MCP, A2A, … Building and Integrating Chaining & Memory ... App Deployment Platform Incorporation The Journey of Developing with Generative/Agentic AI
22. kevindubois Set goals App developer IT operations Data engineer Data scientists ML engineer Gather and prepare data Develop model Integrate models in app dev Model monitoring & management Retrain models Business leadership AI is a team initiative
23. kevindubois Productivity, stability, security
24. kevindubois
25. kevindubois AI/ML + Kubernetes Operators
26. kevindubois openshift_ai_natalevinto.mp4
27. kevindubois
28. kevindubois ▸ AI is overwhelming, but with the right steps it’s not as scary as it seems to get started. ▸ Things are evolving rapidly. ▸ Open Source is key! ▸ Traditional stacks like Java can do AI too ☺ ▸ Cloud Native Computing is more relevant than ever to enable AI development and deployment. Recap
29. kevindubois Thank you! slides opendatahub.io podman - desktop.io docs.quarkiverse.io/quarkus- langchain4j github.com/kdubois/netatmo- java - mcp youtube.com/@thekevindubois linkedin.com/in/kevindubois github.com/kdubois @kevindubois.com @kevindubois@mastodon.social Muchas Gracias!!