SCALABLE AI SEARCH FOR E-COMMERCE: HYBRID RAG ARCHITECTURE AND VECTOR DATABASES

Артем Цимбал; Артем Цимбал

doi:10.30888/2663-5712.2025-34-03-010

SCALABLE AI SEARCH FOR E-COMMERCE: HYBRID RAG ARCHITECTURE AND VECTOR DATABASES

Автор(и)

Артем Цимбал National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” https://orcid.org/0009-0006-8786-8428

DOI:

https://doi.org/10.30888/2663-5712.2025-34-03-010

Ключові слова:

artificial intelligence, hybrid retrieval, retrieval-augmented generation (RAG), vector databases, e-commerce search, embeddings, approximate nearest neighbor (ANN), multi-tenancy, FinOps optimization, latency management, scalable infrastructure

Анотація

The rapid expansion of multi-tenant e-commerce ecosystems with millions of SKUs demands a new generation of scalable, latency-aware, and cost-efficient AI search architectures. Maintaining stable interactive performance (p95/p99 latency) under fluctuating

Downloads

Опубліковано

2025-11-30

Як цитувати

Цимбал, А. (2025). SCALABLE AI SEARCH FOR E-COMMERCE: HYBRID RAG ARCHITECTURE AND VECTOR DATABASES. SWorldJournal, 3(34-03), 3–15. https://doi.org/10.30888/2663-5712.2025-34-03-010