Inspiration
The pharmaceutical market evolves rapidly, with frequent drug launches, regulatory decisions, safety updates, and communication campaigns. However, this information is scattered across news portals, regulatory websites, and official brand channels, making continuous tracking manual and inefficient. We were inspired to build a system that automatically detects and structures market-moving events in real time, enabling pharma teams to stay informed and act faster without relying on delayed or fragmented monitoring.
What it does
The system is an always-on market intelligence agent that continuously monitors public pharmaceutical information sources and identifies key market events such as drug launches, regulatory approvals, safety issues, marketing campaigns, and pricing updates. For each detected event, it:
- Extracts relevant entities (drug, company, indication)
- Classifies the event type using pre-trained AI models
- Converts unstructured text into structured, decision-ready insights
- Displays trends, summaries, and alerts through a live dashboard This allows stakeholders to quickly understand what happened, who was involved, and why it matters.
How will we build it
Data Ingestion The system continuously collects data from public and authoritative sources including pharma news portals, regulatory websites (FDA, EMA, CDSCO), company press releases, and official social media channels such as LinkedIn and X. Event Detection & Entity Extraction Pre-trained NLP and LLM APIs are used to extract entities (drug names, companies) and classify the text into predefined event types. A lightweight rule-based layer acts as a fallback to improve precision for high-impact events. Event Structuring & Storage Each detected event is converted into a structured event object and stored in a relational database for querying, comparison, and trend analysis. Insights & Visualization Aggregated events are analyzed to identify emerging trends, competitor activity, and communication shifts, which are presented through an interactive dashboard with filters and alerts. Always-On Monitoring Scheduled background jobs ensure the pipeline runs continuously, keeping insights up to date without manual intervention.
Built With
- aws-comprehend-medical
- fastapi
- github
- gnews-api
- google-cloud-nlp-api
- javascript
- nextjs
- openai-api
- postgresql
- python
- shadcn
- sql
- tailwindcss
- typescript
- vercel
- x-api
- youtube-data-api
Log in or sign up for Devpost to join the conversation.