Welcome to the central coordination hub for our LLM Zoomcamp team project.
Our vision is to collaboratively build a collection of AI assistants that together represent a simplified Precision Medicine and Drug Discovery platform. While each assistant is an independent LLM Zoomcamp project, they all connect to form a broader healthcare data ecosystem.
We are building specialized AI RAG assistants that map to the standard precision medicine workflow:
Patient ➔ Disease ➔ Gene/Variant ➔ Biomarker ➔ Target ➔ Drug ➔ Clinical Trial ➔ FHIR/mCODE
No single AI can answer all questions accurately. Our multi-assistant architecture ensures high domain accuracy, prevents hallucination, and allows independent scaling.
This GitHub Organization contains three types of repositories:
team-hub: The central brain. Contains our architecture diagrams, team roadmap, datasets registry, and meeting notes.
biomedical-rag-template: The foundational template. Contains our standardized Docker setups, CI/CD pipelines, ChromaDB implementations, and Streamlit UI code. All individual projects must be instantiated from this template.
Each team member owns one of the following independent repositories:
disease-reconciliation-assistantvariant-interpretation-assistantbiomarker-assistantbiomedical-literature-assistanttarget-discovery-assistantdrug-repurposing-assistantclinical-trial-matching-assistantfhir-assistant
| Team Member | Domain / Assistant | Primary Data Sources |
|---|
Step 1: Request access to this GitHub Organization.
Step 2: Navigate to the biomedical-rag-template repository.
Step 3: Click the green "Use this template" button to create your project repository within this organization.
Step 4: Check the Project Kanban Board to pick up your first tasks for the current Zoomcamp week.
Step 5: Read the Data Ingestion Guide before downloading large datasets locally.
- Shared Datasets:
team-hub/datasets_registry/ - Meeting Notes:
team-hub/docs/meeting_notes/ - LLM Zoomcamp: DataTalks.club Course Page