Shaunak Joshi

Back to Projects
ML

MedGet

A drug research platform that brings together FDA clinical data, financial market signals, and AI-powered analysis into a single interface. Built for medical researchers, biotech investors, and curious users who want to explore the lifecycle of FDA-approved drugs — from clinical approval through market performance.

MedGet
PythonReact.jsNext.jsLangChainPostgreSQLFDA API

FDA Data Integration

Pulls real-time data from the FDA's public API — drug approval history, clinical trial outcomes, labeling changes, and adverse event reports — all in one place.

Financial Intelligence

Correlates drug approval events with stock price movements for pharmaceutical companies, giving investors context on how regulatory milestones affect market performance.

AI-Powered Q&A

LangChain pipeline lets users ask natural language questions about any drug — "What were the side effects reported for X?" — and get answers grounded in FDA data.

Research Workflow

Designed for researchers who need to cross-reference publications, approval dates, and financial data without switching between half a dozen different portals.

Technical Highlights

  • LangChain agent orchestrates multi-step FDA data retrieval and summarization
  • PostgreSQL stores normalized drug, trial, and market event data for fast querying
  • Next.js API routes proxy FDA and financial data to avoid CORS and rate limit issues
  • React frontend with real-time search, drug timelines, and comparative views