Hackathon Showcase

Geo-Contextual News–Driven Stock Movement Prediction

Team consisting of a NYU Courant student (IIT Delhi BTech) skilled in JavaScript/React, Node.js, Tailwind, Git; frontend/full‑stack internships and RL robotics experience.

1 member Watch Demo

YouTube Video

Project Description

I started off using a CopilotKit+Langgraph Boilerplate that I revamped later with striking elements for design purposes. I also used Redis for caching my News results in the Store, fetched during each analysis cycle so that Models and Scrappings aren’t rerun and cost me Credits. I also used Tavily extensively to scrape the web in an intelligent manner and fetch me latest News articles of relevance both temporally and geographically so that they can be correlated with appropriate Stock exchanges’ indices. I’ve pushed the latest code to my Github repo and uploaded the Demo to YT.

Prior Work

I started from scratch today morning and wrapped up by 5 or so.

Team

Satyapragnya Kar

Products & Tools

AI Tinkerers CopilotKit FastAPI Langgraph Location API NextJS Redis Sentence Transformers Tavily

Additional Links

https://github.com/SATYAPRAGNYA-KAR/Web-Agents-Hackathon-with-Redis-and-Tavily

Github Repo

Summarizing URL...