Geospatial Disaster Relief Dashboard
Built a Python-based log ingestion pipeline collecting and normalizing public emergency datasets from multiple sources, implementing anomaly detection logic to flag suspicious patterns and produce audit-ready findings Designed automated threat-informed workflows using scheduled Python scripts and LLM-assisted analysis, applying hypothesis-driven investigation techniques to surface irregular activity across ingested data Documented all findings, detection logic, and pipeline configurations in structured formats, producing clear technical outputs suitable for both operational teams and executive review
