Uploaded November 2026 | Updated November 2026, 1 day ago
Learn how to take a user question or instruction, retrieve relevant information from your companyโs datasets, and provide it to the model at generation time. Walk away knowing how to create a search pipeline using a combination of Cohere search models.
To learn more, please check out these resources:
* https://aka.ms/ignite26/next-steps/cloud-ai
๐ฆ๐ฝ๐ฒ๐ฎ๐ธ๐ฒ๐ฟ๐:
* Andy Tran
๐ฆ๐ฒ๐๐๐ถ๐ผ๐ป ๐๐ป๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ถ๐ผ๐ป:
This is one of many sessions from the Microsoft Ignite 2026 event. View even more sessions on-demand and learn about Microsoft Ignite at ignite.microsoft.com
THRSP1066 | Cloud & AI
Theater | (200) Intermediate
#MSIgnite
Chapters:
0:00 - Introduction to intelligent search and retrieval pipelines
00:00:32 - Overview of search pipeline components including parsing, embedding, hybrid search, and re-ranking
00:01:18 - Cohere search models introduced: Parse, Embed, and Re-Rank
00:01:40 - Parse model demo for OCR, visual extraction, charts, tables, and structured document parsing
00:03:34 - Benefits of purpose-built Parse models compared to larger expensive LLMs
00:05:03 - Embed model overview with unified multimodal vector search for text and images
00:06:12 - Retail and e-commerce examples using multimodal search for fashion products
00:07:26 - Manufacturing and technical manual search example using image-based retrieval
00:10:15 - Re-Rank model explained for improving retrieval accuracy and complex query reasoning
00:11:49 - Nike demo showing Re-Rank improving multi-condition product search results
Learn how to take a user question or instruction, retrieve relevant information from your companyโs datasets, and provide it to the model at generation time. Walk away knowing how to create a search pipeline using a combination of Cohere search models.
To learn more, please check out these resources:
* https://aka.ms/ignite26/next-steps/cloud-ai
๐ฆ๐ฝ๐ฒ๐ฎ๐ธ๐ฒ๐ฟ๐:
* Andy Tran
๐ฆ๐ฒ๐๐๐ถ๐ผ๐ป ๐๐ป๐ณ๐ผ๐ฟ๐บ๐ฎ๐๐ถ๐ผ๐ป:
This is one of many sessions from the Microsoft Ignite 2026 event. View even more sessions on-demand and learn about Microsoft Ignite at ignite.microsoft.com
THRSP1066 | Cloud & AI
Theater | (200) Intermediate
#MSIgnite
Chapters:
0:00 - Introduction to intelligent search and retrieval pipelines
00:00:32 - Overview of search pipeline components including parsing, embedding, hybrid search, and re-ranking
00:01:18 - Cohere search models introduced: Parse, Embed, and Re-Rank
00:01:40 - Parse model demo for OCR, visual extraction, charts, tables, and structured document parsing
00:03:34 - Benefits of purpose-built Parse models compared to larger expensive LLMs
00:05:03 - Embed model overview with unified multimodal vector search for text and images
00:06:12 - Retail and e-commerce examples using multimodal search for fashion products
00:07:26 - Manufacturing and technical manual search example using image-based retrieval
00:10:15 - Re-Rank model explained for improving retrieval accuracy and complex query reasoning
00:11:49 - Nike demo showing Re-Rank improving multi-condition product search results










