Authors: Rupesh Kumar Gupta, Udit Sharma, Deepak Yadav, Arjun Singh, Dr. A.P Srivastav, Nitin Kumar Sharma
Abstract: This paper presents an AI‑driven personalization framework within a MERN‑stack Alumni Association Platform and compares three backend stacks—Node.js + MongoDB, Spring Boot + SQL, Django + SQL—for their efficiency in delivering real‑time recommendation microservices. We measure per‑request latency, throughput under concurrent AI inference, and development productivity. Node.js achieves the lowest latency (< 50 ms) and highest throughput for I/O‑bound AI tasks; Spring Boot provides stable CPU‑bound performance with robust scaling; Django offers rapid development at the cost of higher latency. AI personalization boosts event RSVP rates by 35 % and mentorship connections by 28 %. We discuss system architecture, implementation, comparative benchmarks, and implications for technology selection
