Senior Python/Streamlit Developer – Clinical Dashboard & PDF Parser

Project Overview: We are a private longevity and metabolic health clinic launching a structured clinical program. We need an experienced full-stack Python developer to build a secure, internal-only web application that automates our diagnostic workflows. The application must ingest messy medical PDF lab reports, run the extracted data through a pre-defined logic engine, display results on an interactive dashboard with longitudinal tracking and dual-axis charts, and export a clean PDF report. Core Tech Stack: Backend & Parsing: Python, LLM API integration (Claude/Gemini/OpenAI) for unstructured PDF data extraction. Frontend & Dashboard: Streamlit (ideal for rapid, clean internal data dashboards). Charting: Plotly or Matplotlib for dual-axis plots (X-axis: 0, 1, 2 hours; Left Y-axis: Insulin; Right Y-axis: Glucose). Export: WeasyPrint or ReportLab for branded PDF generation. Key Deliverables: 1. Secure PDF Ingestion: A script that parses uploaded medical PDFs (DUTCH, GI-MAP, bloodwork) into a structured JSON format. Rule: Zero local storage (must process in temporary memory to comply with privacy regulations). 2. The Logic Engine & Dashboard: A web interface where clinicians can view patient profiles, historical longitudinal data, and automated phased treatment protocols based on an exact if/then logic tree we will provide. 3. Dual-Axis Charts & PDF Export: Dynamic OGTT curve visualization overlaying historical visits, with a single-click button to export a finalized "Patient Roadmap" PDF report. Requirements: Proven portfolio of deployed Streamlit or Python-based data applications. Experience working with LLM APIs for document parsing. Available to complete the build within 3 to 4 weeks. Screening Questions: 1. Have you built interactive data dashboards with Streamlit or React before? Please link an example. 2. How do you handle secure, temporary-memory processing when parsing PDF files via an API?