APPI Synapse — Document Retrieval Application
Internal RAG application for information retrieval in architecture and construction, built with React, FastAPI, MongoDB, LlamaIndex and LangChain.
What was APPI Synapse?
APPI Synapse was an internal information retrieval tool for the architecture and construction industries. It used retrieval-augmented generation to work with document information.
My role
I developed Synapse during my time as APPI’s co-founder and CTO (2023–2025), integrating the application and document-processing components.
Architecture and technologies
The application combined a React interface, a FastAPI backend and MongoDB. LlamaIndex and LangChain supported document processing for retrieval. This description reflects the application details recorded in my résumé.
Engineering focus
The work connected document processing to an internal application for a specific industry. Synapse and Marketplace were separate applications; the knowledge graph and Pinecone implementation described on the Marketplace page is not assumed to be Synapse’s architecture.