Inside The System
A closer look at how we design, build, and refine intelligent systems that perceive, reason, and act in the real world.
Spatia robotics
· Sep 22, 2026This 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.
Spatia robotics
· Sep 15, 2026Force 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.
Deep Dive
· Sep 7, 2026Why 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.
Spatia robotics
· Sep 1, 2026The Robot Found the Part. Now What?
Why the hardest part of industrial automation starts after perception.
Spatia robotics
· Aug 26, 2026What Happens When Your Second Robot Doesn't Speak the Same Language?
One Perception Engine, Any Robot: The Case for a Robot-Agnostic Abstraction Layer
Robotics
· Aug 26, 2026Zero-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.
Behind the Work
· Aug 25, 2026Beyond 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.
Behind the Work
· Jul 1, 2026Inside The System
A closer look at how we design, build, and refine intelligent systems that perceive, reason, and act in the real world.