Forget RAG Pipelines—Build Production Ready Agents in 15 Mins

on YouTube ↗01h 16m

" summary

Demo builds a working Agentic RAG prototype in ~15 minutes using Contextual AI's managed RAG stack, with notebook, API key and a GUI at app.contextual.ai; it ingests PDFs and tables, does layout/OCR and image captioning, runs hybrid retrieval + a custom instruction-following reranker, then answers multi-document numeric queries with linked attributions and sentence-level groundness checks. Workshop shows API code, unit-test evaluation with a fine-tuned LMU unit judge, and an MCP integration to Claude Desktop; the pipeline is modular, production-ready, and demonstrably handles tens of thousands of documents.

AI-written summary · about

" chapters (44)

1. 0:00Introduction
2. 0:14RAG landscape
3. 3:01Contextual AI platform
4. 4:52Data engineering challenges
5. 5:56Workshop setup
6. 7:31API key management
7. 10:26Ingesting data
8. 12:30Document understanding
9. 14:11Agent architecture
10. 16:49Use cases
11. 18:32Building an agent
12. 20:43Querying agents
13. 23:22Advanced settings
14. 24:50Retrieval components
15. 25:38Groundedness checks
16. 27:32Evaluation pipeline
17. 29:14Evaluation metrics
18. 31:53Extraction deep dive
19. 34:02Query path
20. 34:54Retrieval techniques
21. 37:23Generation strategies
22. 37:57Attribution and fidelity
23. 39:13Groundedness analysis
24. 40:36Unit test criteria
25. 41:02LMUnit performance
26. 41:51Agent improvement
27. 47:16MCP integration
28. 49:00MCP configuration
29. 50:32Pricing and support
30. 52:02Takeaways
31. 53:44User feedback
32. 55:56Data handling
33. 57:38SDK and customization
34. 58:50Cloud deployment
35. 1:00:05Scalability
36. 1:01:45Model repeatability
37. 1:03:03RAG future
38. 1:04:02Domain specificity
39. 1:05:34Document parsing
40. 1:06:35Data ingestion
41. 1:07:48Website sync
42. 1:08:44Entitlements and security
43. 1:10:07Fact verification
44. 1:12:34Closing

" related talks

• 01h 13mBuilding Production-Ready AI Systems in 90 MinutesNick Tchayka · PyData NYC 2024
• 41mFrom RAGs to riches: Build an AI document interrogation app in 30 minsEswaramoorthy, Pothina · PyData NYC 2023
• 27mFrom RAGs to riches: Build an AI document interrogation app in 30 minsPhilip Meier · PyData Global 2023
• 21mHow RAG is Redefining AI CapabilitiesDina Helmy · DSC MENA 2025
• 39mAgentic RAG Platform for AI-Assisted WorkflowsMichael Marolda, Drew Lent, Quinton Buesching · Progress Data Platform Summit 2025
• 18mLLMs and GenAI: A Guide to Building Your Own RAG Model for Document InteractionPaula Oses · RTC.ON 2024
• 25mSet Up an Open Source RAG in 30 MinutesPietro Mele · Devoxx France 2024
• 12mCodeless Generative AI Pipelines: Prompt AutomationTimothy J Spann · Conf42 Prompt Engineering 2024