Real-time chat with LLM-based systems
Designed and built the real-time chat interfaces people use to work with SWS's LLM-based systems.
An AI-powered platform for real-time conversational and analytical workflows.
I lead frontend development for SWS at John Snow Labs: real-time AI chat interfaces, the component library underneath them, and product modules including Patient Journey, Knowledge Base and Vector Database integrations.
A custom component library in React, TypeScript, Tailwind CSS and Ant Design keeps the UI consistent across modules. Server state goes through React Query. On the backend I contribute NestJS REST APIs for chat workflows and knowledge retrieval, working closely with the AI and backend teams to ship features end to end.
Designed and built the real-time chat interfaces people use to work with SWS's LLM-based systems.
Built and scaled a custom component library on Tailwind CSS and Ant Design, improving UI consistency and development speed, with reusable components whose server state is handled through React Query.
Delivered key modules including Patient Journey, Knowledge Base and Vector Database integrations, and kept UX patterns consistent across them.
Contributed NestJS REST APIs for chat workflows and knowledge retrieval, then integrated them on the frontend with attention to data contracts, error handling and performance.