This project captures live driving data from the racing simulation game Assetto Corsa (AC) and streams it to a Kafka message broker for downstream processing in a data pipeline. It serves as a test system for building, validating, and troubleshooting data pipelines using realistic and high-frequency sensor data.

This repository contains a multi-agentic system that automatically builds a structured cricket dataset by extracting live match commentary from web sources at regular intervals. It converts unstructured updates into ball-level records and dynamically evolves the schema by detecting new attributes and adding columns, enabling continuous data flow