We are building thinking systems.
Moving beyond artificial prediction to design intelligence that perceives, reasons, and acts.
Intelligence is not magic.
It is structure.
The world has enough AI that predicts. We are interested in AI that understands. We study how humans see, decide, adapt, and learn then we design machines that do the same. At Docilis Labs, we are not chasing trends. We are building thinking systems.
Perception
Absorbing and interpreting raw sensory environments via real-time spatial arrays and edge visual transformers.
Reasoning
Temporal state estimation, hazard trajectory prediction, and adaptive contextual understanding.
Action
Executing deliberate physical maneuvers or sub-millisecond tactile and audio actuation in human spaces.
"Sometimes that action is software. Sometimes it is a robot.
Always, it is intentional."
Docilis Labs was founded from a simple but uncomfortable question: Why do machines compute perfectly, yet understand nothing?
While most AI systems optimize predictions, Docilis Labs exists to design intelligence systems that reason, adapt, and perceive the world the way humans do. It is not a software factory. It is a research-first lab where theory meets prototypes, and prototypes evolve into embodied machines.
"Africa will not consume intelligence forever. We will create it."
Applied Visual Intelligence.
Docilis Visual Intelligence Engine (DVIE) v1
Project AI4Sight is officially in motion. DVIE v1 is operating in the real world as the core cognitive vision system behind AI4Sight - an assistive wearable hardware designed to support visually impaired school children in Kenya.
Giving intelligence a physical form.
The natural evolution of the Docilis Visual Intelligence Engine is embodiment. Our active research roadmap focuses on translating cognitive software architectures into physical hardware platforms.
Docilis Mini
Status: Active R&DA compact humanoid research platform designed to test cognitive architectures, spatial reasoning, and bipedal kinematics in human environments.
Docilis Rover
Status: ConceptualAn autonomous wheeled platform focused on rugged terrain navigation, extended sensor arrays, and persistent environmental mapping.
From the Workbench.
Unfiltered engineering postmortems, mathematical derivations, and hardware logs directly from the research lab.
Announcing DVIE v1.5 - Real-World Intelligence Beyond the Cloud
Today, we are announcing Docilis Visual Intelligence Engine (DVIE) v1.5, the software architecture powering Project AI4Sight. Here is how we engineered smart interruption audio queues, 5-zone spatial grids, and zero-cloud privacy.
Lab Notes - Teaching AI Object Permanence & Motion
Picture setting your mug down next to your laptop. Your brain recognizes both are still. Here is how we solved the Ping-Pong oscillation bug and camera jitter on edge vision hardware.
Lab Notes - Squeezing Heavy AI onto a $100 Chip
Building high-tier AI software inside a cloud lab with unlimited compute is one thing. Making it run on a $100 credit-card-sized board on smart glasses is a completely different engineering challenge.