| Backend API Server |
FastAPI (Python) |
Async APIs, authentication, ingestion, timeline, and chat services |
| Mobile Application |
React Native |
Cross-platform patient-facing mobile application |
| Workflow Orchestration |
LangGraph |
Stateful AI workflow orchestration and routing |
| AI Framework |
LangChain |
LLM integrations, retrieval, embeddings, and document pipelines |
| OCR & Document Parsing |
PyMuPDF + Tesseract OCR |
Extract text from prescriptions, reports, and scanned documents |
| Local LLM Server |
Ollama |
Self-hosted model serving and inference |
| Chat & Extraction Model |
Qwen 2.5 3B |
Medical information extraction, intent routing, summarization, and RAG responses |
| Embedding Model |
mxbai-embed-large |
High-quality semantic embeddings for medical document retrieval |
| Relational Database |
PostgreSQL (Neon) |
Structured patient timeline, diagnoses, medications, and metadata |
| Vector Database |
Qdrant Cloud |
Semantic search and retrieval over medical memories |
| ORM & Database Client |
Prisma |
Type-safe PostgreSQL access and schema management |
| Containerization |
Docker & Docker Compose |
Local development and deployment consistency |
| API Documentation |
OpenAPI / Swagger |
Interactive API exploration and testing |