Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM67718
Volume:
10 |
10 |
Oct 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
MCA School of Sciences
G H Raisoni International Skill Tech University Pune
School of Sciences
G H Raisoni International Skill Tech University Pune
School of Sciences
G H Raisoni International Skill Tech University Pune
School of Sciences
G H Raisoni International Skill Tech University Pune
MCA School of Sciences
G H Raisoni International Skill Tech University Pune
MCA School of Sciences
G H Raisoni International Skill Tech University Pune
Conventional digital commerce platforms offer essential keyword search, static category filters, product recommendations, and customer review listings; however, consumers continue to experience significant cognitive overhead when evaluating complex requirements, comparing technical specifications across multiple browser tabs, and distilling vast volumes of user feedback. Furthermore, traditional static chatbots fail to execute actionable backend operations, while granting artificial intelligence (AI) models direct database connections introduces severe security and transactional vulnerabilities. This paper presents an AI-powered multi-agent e-commerce platform designed to bridge conventional web stores with an autonomous multi-agent intelligence layer. The architecture employs Next.js with TypeScript for the responsive frontend, Node.js with Express.js for core transactional services, and Python with FastAPI for high-performance AI microservices. A central agent orchestrator interprets natural language user intents and delegates tasks to five specialized agents: Shopping Agent, Recommendation Agent, Comparison Agent, Review Agent, and Support Agent. Grounded state retrieval and actionable operations are enforced through Retrieval-Augmented Generation (RAG) using pgvector semantic search and strict, tool-calling REST APIs exposed by the Express backend. This paper details the system design, agent communication boundaries, data storage strategy, and experimental performance results, demonstrating a scalable, secure, and highly responsive framework for next-generation intelligent e-commerce.
KEYWORDS
This paper presented an AI-powered multi-agent e-commerce platform that successfully bridges conventional digital retail web applications with an autonomous multi-agent intelligence layer [1]. By segregating user requests among five specialized agents (Shopping, Recommendation, Comparison, Review, Support) and strictly enforcing tool-calling boundaries via validated Express.js REST APIs, the architecture resolves cognitive shopping fatigue while establishing robust database security. Experimental evaluations demonstrate 96.4% intent routing accuracy and an average end-to-end response latency of 380 ms.
[1] P. Rathod and T. Rakshe, "AI-Powered Multi-Agent E-Commerce Platform Synopsis," Dept. MCA, G H Raisoni Int. Skill Tech University., Pune, India, Project Synopsis, 2026. [2] P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, H. Küttler, M. Lewis, W. Yih, T. Rocktäschel, S. Riedel, and D. Kiela, "Retrieval-augmented generation for knowledge-intensive NLP tasks," in Proc. Advances in Neural Information Processing Systems (NeurIPS), vol. 33, pp. 9459–9474, 2020. Link [3] OpenJS Foundation, "Express.js Infrastructure Guide and Security Best Practices," 2023. [Online]. Available: https://expressjs.com [4] S. Ramírez, "FastAPI Framework for High-Performance Python Microservices," 2023. [Online]. Available: https://fastapi.tiangolo.com [5] PostgreSQL Global Development Group, "pgvector: Open-Source Vector Similarity Search for PostgreSQL," 2024. [Online]. Available: https://github.com/pgvector/pgvector [6] Vercel, "Next.js Documentation: App Router and Server Components Architecture," 2024. [Online]. Available: https://nextjs.org/docs [7] H. Chase, "LangGraph: Building Stateful, Multi-Actor Applications with LLMs," LangChain Inc., 2024. [Online]. Available: https://python.langchain.com/docs/langgraph [8] X. Zhang, Y. Wang, and L. Chen, "Multi-Agent Systems for Autonomous E-Commerce and Recommendation Pipelines," IEEE Transactions on Knowledge and Data Engineering, vol. 36, no. 4, pp. 1820–1834,Apr. 2024. [Online] Available: https://doi.org/10.1007/s11704-024-40231-1 [9] E. Siddiqui, V. Terikuti, and X. Lu, "LLM-based Semantic Search for Conversational Queries in E-commerce," arXiv preprint arXiv:2601.16492, 2026. [Online] Available : https://arxiv.org/pdf/2601.16492 [10] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, "Attention is all you need," in Proc. Advances in Neural Information Processing Systems (NeurIPS), vol. 30, pp. 5998–6008, 2017. [Online] https://arxiv.org/abs/1706.03762
Publication Date
10/7/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
8
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...