arxiv-research Next.js • Supabase • Vercel AI SDK • LangGraph • FastAPI ●​ Built a full-stack AI-powered research tool that lets users search, store, and interact with arXiv papers in custom workspaces. ●​ Developed an AI chat interface that allows users to converse with papers using document-aware memory and streaming responses. ●​ Designed a LangGraph-based multi-agent system that analyzes papers via agent-led question–answer cycles. ●​ Integrated Supabase for real-time database operations and authentication. ●​ Deployed on Vercel with FastAPI as the orchestration backend, enabling edge-optimized performance and scalable AI interaction.