A1-3 PRIVATE (test) AI stack deploys

References

LOCAL/REMOTE LLM = brains
LOCAL Agents bring the deterministic workflow control
- Agents are many types and complex
- usually only change API call to connect to model (model is just the "brains"?)
- agent is probably tightly designed for the RAG, tools, ets
LOCAL UI is not so important, but put it in here because it will usually connect to agent
LOCAL
- RAG
- tools
REMOTE
- MCP to external tools
- MCP/A2A to other agents

#380_(OK)_research_agent_pydanticai_WITTEVEEN_. PydanticAI COLAB.
#377_()_examples_PYDANTIC_
#373_(OK)_a2a+mcp_GERMAN_
#371_(OK)_agents_pydanticai_EBBELAAR_
#369_()_ai_cookbook_mcp_EBBELAAR_. Excellent demo of MCP. Contains the client side implementation that I having been wanting to implement, but this is the first demo I have found that has the detail I needed.
#368_()_groq_cookbook_GROQ_. Going through the Groq cookbook. My lab notes in this doc show that running AI without programming skills isn't going to happen anytime soon.

(other)

#498b_ai_deep_dive_KARPATHY_v09_25.0518.docx ????
#421_()_platform_docs_OPENAI_v01_25.0120.docx junk?? (assistants)
#420_()_gemini_api_qs_GOOGLE_v01_25.0120.docx
#389_()_hugface_demos_GPT_v01_25.0515 9 pages , 7 steps, gradio, inference failed
(DEPLOY) #385D_(BINGO)_deploy_openai_agents_sdk_GPT_v01_25.0513 (1)

(Local LLM (HugFace))

#492_hface_course_HFACE_v02_25.0519
(WSL ok, WIN11 fail) #390_(BINGO)_gpt2_in_wsl2_local_GPT_WITTEVEEN_v03_25.0518.docx