Applied AI
Putting machine learning to work in real operations.
- AI integration for enterprise systems
- Data pipelines and model training
- Model evaluation and deployment
- Reinforcement learning
- Sim-to-real transfer
- Learning from demonstration
Applied AI and robotics, from prototype to production.
Dexteris AI is a Detroit engineering firm that designs, integrates, and deploys applied AI, from learning pipelines and vision systems to the robots and industrial lines that put them to work.
Platforms we work with
- Unitree G1
- Unitree Go2
- Boston Dynamics Spot
- Universal Robots UR3e
- Universal Robots UR10e
What we do
Vision & Perception
Giving the robot an accurate picture of its workspace.
- 3D perception and reconstruction
- 6-DoF pose estimation
- Detection and segmentation
- Hand-eye calibration
- Sensor fusion
- Obstacle avoidance
Robotics & Motion Control
Getting hardware running, and running safely.
- Hardware diagnostics and bring-up
- SDK and driver integration
- Whole-body control
- Motion planning
- Teleoperation and XR interfaces
- ROS 2 integration
Industrial Integration
Connecting robots to the line and the people who run it.
- Robot cell integration
- PLC and line interfacing
- Machine vision inspection
- Safety interlocks and watchdogs
- Operator interfaces
- Commissioning and training
Case Study
Teleoperation and learned manipulation on the Unitree G1
For a Detroit venture studio, we built real-time control of a G1 humanoid from a Meta Quest headset or a web browser, with obstacle avoidance and whole-body control running underneath. We also built the reinforcement-learning pipeline that trains its manipulation policies in simulation, and verified the transfer on the physical robot.
- Client
- Venture studio, Detroit
- Year
- 2026
- Platform
- Unitree G1
About Dexteris AI
Founded in Michigan, Dexteris AI builds AI-driven robotics, automation, and enterprise systems that produce measurable results in real operations. Clients work directly with the engineers writing the code, and our ties to university research keep our methods current.
Leadership
Blaine Oania
Principal Engineer
Blaine specializes in applied AI, from embodied intelligence on robots to industrial enterprise systems, with a background spanning controls design, industrial software, and robotics.
- 5+ years in robotics, automation, and manufacturing
- M.S. Robotics & AI
- B.S. Computer Science
- Speaker, Michigan DevFest 2024: “How to Train Your Robot!”
Tell us about your project
Tell us what you're working with and what you need it to do. We'll reply with an initial assessment and suggested next steps.
- Location
- Detroit, Michigan