Built a full-stack SaaS platform

Built a full-stack SaaS platform enabling businesses to create branded, knowledge-backed chatbots with an embeddable JavaScript widget deployable on any website with a single script tag. Engineered a complete RAG ingestion pipeline supporting PDFs, URLs, and raw text — chunking, embedding via FastEmbed, and storing dense and sparse vectors in Qdrant for hybrid semantic retrieval. Designed multi-tenant system architecture with onboarding flows, bot CRUD, per-bot analytics dashboard (message events, top questions, routing breakdown), and an admin moderation panel. Containerised the full application using Docker with a multi-stage build; deployed on Railway with a single-process SPA+API pattern and automated health check endpoints. Tech: React, Vite, TypeScript, FastAPI, MongoDB, Qdrant, Groq, Docker, Railway, Tailwind CSS, shadcn/ui. Link: https://dashbot-production-ff88.up.railway.app