- 🛰️ Geospatial Data Scientist turning LiDAR, satellite, and hazard data into decision-ready intelligence
- 🤖 Building agentic AI systems — LLM copilots and MCP servers over spatial & hazard APIs
- ☁️ Designing enterprise data platforms on AWS — CDK, microservices, PostGIS at scale
- 🧠 Deep in computer vision & inference — ONNX pipelines, PDAL, production-grade CV
- 🌱 Currently shipping: agentic geospatial copilots and real-time hazard-exposure pipelines
graph LR
A["🤖 ML & AI"] -->|inference| B["🛰️ Geospatial"]
B -->|data| C["☁️ Cloud"]
C -->|ops| A
Auto-updated daily from my most recently active repos — see update-projects.yml.
| Project | Description |
|---|---|
| SurgeExposure | A cloud-native pipeline overlaying NOAA storm-surge and flood-inundation data against Overture Maps building footprints, scoring per-asset flood exposure across 8 coastal regions — with a map and API. |
| fflood-nep | A reproducible flood-response pipeline for Nepal's August 2026 Rasuwa/Bhote Koshi flash flood — runs Sentinel-1 SAR change detection for flood extent, scores building/road/facility exposure by munici… |
| wildfire-copilot | A LangGraph agent for utility wildfire risk — combines a trained XGBoost susceptibility model with live NASA/NWS hazard feeds and RAG over real wildfire-mitigation filings to answer plain-English ris… |
| geomlint | A CI-first data-quality linter for vector geospatial files — catches invalid geometry, CRS confusion, and topology errors that mainstream data-observability tools skip. No GDAL required; pip install… |
| Geospatial Data Copilot | Ask a plain-English question about San Francisco's utility infrastructure — a local LLM agent (LangGraph ReAct, Ollama) translates it to spatial SQL, runs it, and answers with a summary and interacti… |
| HazardMCP | An MCP server unifying FEMA, USGS, NOAA, NWS, and NASA hazard data behind a single AI-agent interface — one-call location risk summaries and map rendering, compatible with Claude Desktop. |
