Sorry, you need to enable JavaScript to visit this website.

SNIA Developer Conference September 15-17, 2025 | Santa Clara, CA

Staff Engineer

Samsung Semiconductor India Research, Bangalore

Roshan Nair is a Staff Engineer at Samsung Semiconductor India Research where he works in the Global Open-ecoSystem Team (GOST). His current work focuses on research and collaboration on open ecosystem enablement of emerging technologies like CXl and NVMe FDP. He has rich experience in the industry developing and researching distributed storage systems, workload analysis, ML for systems and host software solutions. He has 5 patents and patent applications in the field.

Towards Memory Efficient RAG Pipelines with CXL Technology

Submitted by diegonika on

Various stages in the RAG pipeline of AI Inference involve large amounts of data being processed. Specifically, the preparation of data to create vector embeddings and the subsequent insertion into a Vector DB requires a large amount of transient memory consumption. Furthermore, the search phase of a RAG pipeline, depending on the sizes of the index trees, parallel queries, etc.

Subscribe to Roshan Nair