Episode publish date
August 17, 2025 2:00 PM (UTC)
Last edit date
Aug 29, 2025 1:30 AM
Last snip date
August 29, 2025 2:29 AM (GMT+1)
Last sync date
August 29, 2025 2:30 AM (GMT+1)
Show
MLOps.community
Show notes link
Snips
13
Warning
⚠️ Any content within the episode information, snip blocks might be updated or overwritten by Snipd in a future sync. Add your edits or additional notes outside these blocks to keep them safe.
‣
Your snips
‣
[01:02] Fact-Based Agent Learns Over Time
‣
[03:48] Let Audience Dictate Agent Intervention
‣
[06:09] Connect Sources To A Product Ontology
‣
[08:39] Terms Vary By Context; Agents Need Synonyms
‣
[12:21] Be Concise And Ground Responses
‣
[15:25] Stage Automation With Human Approval
‣
[21:35] Docs Should Be Dense And Code-Linked For Agents
‣
[25:09] Make Saving Truth Easier Than Editing Docs
‣
[31:06] Agents Should Share Curated Knowledge
‣
[38:10] Use Delta-Based Evals For Living Knowledge
‣
[44:53] Prioritize Core Integrations, Handle Long Tail Auth
‣
[48:44] Design For A Living Single Source Of Truth
‣