Mdical_shift

this is the backend connection/backend for our project (swagger ui)

Inspiration The pharmaceutical industry generates an overwhelming volume of news, announcements, and regulatory updates every day. While information is abundant, identifying what actually matters—especially early signals of market change—remains a challenge. We were inspired by the gap between raw information availability and actionable intelligence, particularly for strategy, regulatory, and competitive teams that need clarity rather than noise. What it does Medical_shift is an always-on pharma market intelligence system that continuously monitors industry communications, detects meaningful changes in activity, and transforms them into concise, explainable insights. Instead of summarizing news, the system identifies abnormal patterns and surfaces only what requires attention. How we built it We designed a modular backend that ingests pharma news sources, applies deterministic change-detection logic to identify signals, and converts those signals into structured insights. An AI layer enhances clarity by refining explanations without influencing decision logic. A minimal frontend presents insights as executive-ready tiles for rapid understanding. Challenges we ran into Key challenges included filtering noise without overfitting, ensuring explainability in AI usage, and building a reliable end-to-end pipeline under hackathon time constraints. Accomplishments that we're proud of We successfully built a fully working intelligence pipeline that is explainable, conservative, and demo-ready, with real-time data flow and a clean user interface. What we learned We learned the importance of separating signal detection from AI reasoning, prioritizing trust, transparency, and usability over complexity. What's next for Medical_shift Future plans include richer signal types, historical trend analysis, personalization for pharma roles, and deployment as a scalable enterprise intelligence platform.