Company
Arden Meridian
Analyst-reviewed property intelligence for commercial teams — satellite imagery, geospatial data, and public records turned into prioritized answers.
Founder & CEO · Arden Meridian
Zander Polk
I run Arden Meridian, turning satellite imagery, geospatial data and public records into property intelligence commercial teams can act on. Separately I build BEACON, a research prototype for satellite conjunction triage. I have been writing software since 2016.
The Company
Arden Meridian
I founded Arden Meridian in 2026 to answer a narrow and expensive question for commercial teams: out of everything in this territory, which properties are actually worth your time?
Arden Meridian
Founder & CEO Founded 2026Overhead insight. Ground-level decisions.
We produce analyst-reviewed property intelligence — combining satellite imagery, geospatial data, and public records into reports that tell commercial teams which properties are worth their time, and why. The difference is where the work stops: most tools hand over a dataset, we deliver finished research.
ardenmeridian.comWhere it’s applied
Commercial Solar Prospecting
Solar Maintenance & Repair
Building Maintenance Intelligence
Distributed Energy Siting
Property Risk & Resilience
Industrial & Data-Center Siting
Custom Territory Intelligence
About
Why I do all three
I run a company, do research, and still write code. The three keep each other honest: research shows me what’s actually possible, engineering shows me what it costs, and running the business shows me whether anyone cares.
The through-line is calibrated judgment. I build systems that are honest about how much they know — whether that’s a conjunction model reporting its own uncertainty or a property report saying which findings are solid and which aren’t.
Charlotte, North Carolina
Read the full backgroundSelected Work
What I’m building
A company, a set of research prototypes, and a decade of game and platform work.
Research prototype
BEACON
Calibrated, uncertainty-aware machine learning for satellite conjunction triage on public CDM data. Open source, archived on Zenodo.
Library
TaskRuntime
A structured asynchronous task and lifecycle library for Polytoria Luau — cancellation, retries, bounded concurrency, and deterministic test scheduling.
Research
Research and open-source work
My research sits where space systems meet machine learning. The interesting question is rarely whether a model can predict something — it’s how much anyone should trust the prediction when they have to act on it.
01
Space Systems
Orbital dynamics, space-domain awareness, and Earth observation — how objects and sensors actually behave on orbit, and what that constrains about anything inferred from the data they return.
02
Machine Learning
Applied ML where being accurate isn’t sufficient — calibration, uncertainty quantification, and rare-event prediction. A model that says ninety percent should be right ninety percent of the time.
Start a conversation
Whether you’re working on geospatial problems, interested in what we’re building at Arden Meridian, or want to compare notes on space systems and machine learning — I’m glad to have the first conversation.