In this episode of the Spatial Stack, Matt sits down with Kyle Walker, Professor of Geography at TCU and the creator of popular R packages like tigris and tidycensus.

Kyle dives into why he views US Census data as critical infrastructure and how open data is fundamentally transforming decision-making across industries like real estate and energy. He shares the origin story of his open-source work, explaining why he champions the R programming language for full-stack geospatial analysis. The conversation also covers the evolution of web mapping, from the laborious process of rendering dot-density maps to the blazing-fast performance of modern tools like PMTiles.

Finally, Kyle reveals how generative AI specifically Claude Code and the Zed editor is serving as his ultimate coding assistant, allowing him to rapidly build complex projects like the mapgl package and turn his ideas into reality faster than ever.

Connect with Kyle:

X/Twitter: https://x.com/kyle_e_walker
LinkedIn: https://www.linkedin.com/in/walkerke/
Bluesky: https://bsky.app/profile/kylewalker.bsky.social

00:01:00 – Welcome and Kyle Walker’s Background at TCU
00:06:18 – Why US Open Data is Critical Infrastructure
00:09:20 – Demystifying Census Data with tigris and tidycensus
00:15:48 – Applied Spatial Data: Real Estate and Forecasting Models
00:18:28 – The Evolution of High-Resolution Dot Density Maps
00:23:48 – The Human Element: How People React to Seeing Data Maps
00:29:14 – R vs. Python: Why R is a Geospatial Powerhouse
00:37:44 – Accelerating Development: Using Claude and AI for Coding
00:43:40 – The Future of Mapping: PMTiles, Segment Anything, and LLMs
00:48:18 – Where to Find Kyle’s Book, Tools, and Workshops

---

🚀 Join The Spatial Lab:
Stop guessing at your career path. Get direct mentorship, advanced training, and a roadmap to these high-value roles inside The Spatial Lab.
👉 https://forrest.nyc/spatial-lab/

📰 Daily modern GIS insights: https://forrest.nyc

CONNECT WITH ME
📸 Instagram: https://www.instagram.com/matt_forrest/
💼 LinkedIn: https://www.linkedin.com/in/mbforr/
📧 Newsletter: https://forrest.nyc
🌐 Website: https://forrest.nyc

We have never had more data about our planet: petabytes of satellite imagery, aerial photos, and sensor readings collected daily. Yet, turning that massive volume of "noise" into a clear signal remains the fundamental challenge of the geospatial industry.

In this episode of the Spatial Stack, I sit down with the engineering and product minds from Wherobots: Ryan, Phil, and Len - to tear down the architecture required to handle Earth Observation data at a planetary scale. We move beyond the buzzwords to discuss the engineering "war stories" of building resilient inference pipelines.

We dive deep into why the industry is moving away from simple computer vision toward "Large Earth Models" that function like LLMs for the physical world. We also get into the weeds of the tech stack: the battle between Dask and Ray for distributed compute, why Cloud-Optimized GeoTIFFs (COGs) aren't always the answer for inference, and how formats like Zarr are unlocking multidimensional analysis.

In this episode, we cover:

The Data Bottleneck: Why "garbage in, garbage out" is still the biggest hurdle in monitoring a changing planet.

Infrastructure Realities: The specific limitations of Google Earth Engine and why we needed a cloud-agnostic approach.

Engineering Pivot: Why Wherobots migrated from Dask to Ray to solve "crashing cluster" syndromes and memory management issues.

The Future of GeoAI: How embeddings and foundation models are compressing petabytes of data into searchable, semantic insights.

✅ Sign Up for Wherobots: https://wherobots.com/
✅ Learn more about Apache Sedona: https://wherobots.com/apache-sedona/
✅ Learn more about RasterFlow: https://wherobots.com/blog/rasterflow-earth-observation-inference-engine/
✅ Sign Up for the RasterFlow Private Preview: https://wherobots.com/rasterflow-preview/

00:00 – Teaser: The "Garbage In, Garbage Out" problem in GeoAI
00:01:51 – Introductions & Icebreakers (The controversial ice cream opinions)
00:03:08 – The Challenge: Monitoring a changing Earth at scale
00:10:30 – Data Engineering: The hidden complexity of NAIP, clouds, and tiling artifacts
00:14:19 – Modeling Reality: Why Computer Vision models fail on geospatial data
00:21:51 – The Google Earth Engine Debate: Walled gardens vs. bringing compute to the data
00:27:53 – Introducing Rasterflow: A new architecture for scalable inference
00:36:51 – The Engineering Story: Why we switched from Dask to Ray
00:43:40 – File Formats: Why Zarr is superior to COGs for multidimensional inference
00:47:40 – Workflow Walkthrough: Running the "Fields of the World" model
00:51:40 – Embeddings, Foundation Models, and Large Earth Models
00:57:40 – How to get started with Rasterflow

📰 Modern GIS insights: https://forrest.nyc

CONNECT WITH ME
📸 Instagram: https://www.instagram.com/matt_forrest/
💼 LinkedIn: https://www.linkedin.com/in/mbforr/

🌐 Website: https://forrest.nyc

There are trillions of dollars invested in the physical world every da: infrastructure, supply chains, and our planet.

Yet many of these massive decisions are made without the data to back them up. For too long, geospatial analytics has been gated behind specialized teams and siloed technology, treated as "spatial is special" rather than just another data type.

In this episode, we sit down with Damian Wiley from Wherobots to break down how cloud architecture is finally closing this gap. With a heavy-hitting background from AWS EC2 and Databricks, Damian explains the shift from transactional databases to the Lakehouse architecture and why "Zero ETL" is the holy grail for data engineering.

We dive deep into why spatial data shouldn't be gated, how open table formats like Iceberg are changing the game, and why the future involves AI agents that can directly query the physical world.

If you are a data engineer, developer, or leader looking to unlock location intelligence without the headache of complex infrastructure, this conversation is for you.

✅ Sign Up for Wherobots: https://wherobots.com/
✅ Learn more about Apache Sedona: https://wherobots.com/apache-sedona/
✅ What is Apache Sedona: https://wherobots.com/blog/what-is-apache-sedona/
✅ Test out SedonaDB: https://sedona.apache.org/sedonadb/latest/
✅ Connect with Jia on LinkedIn: https://www.linkedin.com/in/wyliedamian/

00:00 - The Trillion Dollar Data Gap: Investing in the physical world without intelligence
02:15 - From AWS EC2 to Geospatial: Damian’s journey from cloud infrastructure to spatial data
06:40 - "Spatial is Special" No More: Breaking down silos and making spatial data "just data"
09:00 - The Lakehouse Advantage: Decoupling storage and compute for economic agility
12:15 - Fragmented History: Why geospatial tech became so compartmentalized
17:30 - Real-World Impact: Optimizing supply chains and climate response with frequent data
22:45 - The Economics of Analytics: Lowering the Total Cost of Ownership (TCO) for pipelines
28:30 - AI Agents & The Physical World: Connecting LLMs to ground-truth reality
37:00 - Compute Strategy: When to use OLAP vs. OLTP for spatial workloads
46:00 - Zero ETL & The Future: How Iceberg and open standards enable interoperability
51:20 - Getting Started with SedonaDB: Vibe coding and the future of spatial queries

📰 Daily modern GIS insights: https://forrest.nyc

CONNECT WITH ME
📸 Instagram: https://www.instagram.com/matt_forrest/
💼 LinkedIn: https://www.linkedin.com/in/mbforr/
📧 Newsletter: https://forrest.nyc
🌐 Website: https://forrest.nyc