Ideas worth understanding.

Stories, deep dives, and insights from the team building spatial intelligence for the physical world.

In-Context Learning for Robots
· ~4-5 min read

This AI Watched One Human Video and Suddenly Knew How to Use Its Hands

Every robot deployed today lives with the same limitation: it only knows what it was trained on. Change the object, change the lighting, change the task slightly, and performance can fall apart. The traditional fix - collect more data, retrain, redeploy - doesn't scale to the messiness of the real world.

Read article
spatia spatial intelligence  Robot-Agnostic Perception
· ~5 min read

Force Control: The Feature That Actually Solves the Last Inch

Pose estimation gets a robot to the part - force control is what lets it feel what happens next. This post breaks down how Flexiv, Franka Emika, KUKA, Universal Robots, and ABB/Doosan each implement force sensing, and why that sensing is the missing half of the last-inch problem.

Read article
Spatia
· 6 min read

Why We Didn’t Build on a VLA (And When You Actually Should)

In Physical AI, the best architecture isn't always the most sophisticated one. This article explains why Spatia chose a modular approach for industrial robotics instead of building around a VLA, and when a VLA architecture can actually be the better choice.

Read article
The Last Inch Problem in Robotic Manipulation
· 6 min read

The Robot Found the Part. Now What?

Why the hardest part of industrial automation starts after perception.

Read article
spatia spatial intelligence  Robot-Agnostic Perception
· 7 mins

What Happens When Your Second Robot Doesn't Speak the Same Language?

One Perception Engine, Any Robot: The Case for a Robot-Agnostic Abstraction Layer

Read article
Zero-shot
· 8

Zero-Shot 6D Pose Estimation: The End of the Re-Teach Tax?

A new part used to mean a new perception problem — collect data, label, train, validate, tune, deploy. Then do it again when the SKU changes. That workflow is starting to break.

Read article
dvanced robotics system in operation
· 8 min read

Beyond Perception: Building Machines That Understand the World

Autonomous systems don't operate in static environments. They must continuously interpret change, understand context, and make decisions in the physical world. This is how we think about building intelligence that goes beyond perception.

Read article
Modern warehouse interior with automated systems
· 7 min read

Physical AI in Warehouses

How autonomous systems are transforming warehouse operations and why spatial understanding is the key to making it work at scale.

Read article
Advanced robotics system in operation
· 8 min read

Inside The System

A closer look at how we design, build, and refine intelligent systems that perceive, reason, and act in the real world.

Read article
Team discussion in a modern research facility
· 5 min read

Conversations

We bring together people who ask better questions, challenge assumptions, and believe that understanding is the first step toward building what matters.

Read article

Want to learn more about spatial intelligence?

Follow our work as we build systems that help machines understand the world around them.

Get in touch