AI Research & Engineering

RetgenAI

Engineering the systems that turn raw data into actionable intelligence.

Read the blog

Latest Writing

Writing on geospatial, data engineering, RAG and beyond.

View all posts

06.19.2026Research3 min

Why Binary Change Detection Isn't Enough

06.12.2026Use Case3 min

Zero-Shot Damage Assessment: Evaluating MBCTD on the 2025 LA Wildfires

05.21.2026Release3 min

MBCTD: Open multi-label building change type detection at the pixel level

04.06.2026Engineering12 min

What is Hybrid Search?

Inspection View

I built a custom web app to test my MBCTD model on the 2025 LA wildfires. Here's a look at how it generalized to real-world disaster damage on the fly.

I'm releasing MBCTD, a deep learning model that detects demolished, new, and unchanged buildings independently per pixel, including replacement sites that single-label models can't represent.

"Why Binary Change Detection Isn't Enough"

Binary change detection tells you something changed. Directional multi-label detection tells you what happened. Here's why that distinction matters for urban monitoring and damage assessment.

CategoryResearch

Metrics3 MIN

Let's build something impactful.

I engineer data pipelines and AI systems that turn raw information into actionable insights.

Get in touch