Robotics + AI Classes for Teens
RoboLab teaches teenagers how to build and program real robots — combining hands-on hardware with modern AI. Ages 13–17. Expert-led. No experience needed.
The final project isn't a competition or a demo day. Students go out into the world, find one person with a real problem, build a robotics or AI solution, and deliver it to them. The bar is simple: does it work for a real human?
Lessons are adapted to each student's interests. These are the projects in our library — your child works through the ones that fit them.
Full CAD design in Fusion 360, 3D printed, assembled by hand. Students specify tolerances, pick fasteners, and revise based on load testing.
IR sensor array + PID control loop. Tracks a taped line at increasing speeds. First real closed-loop control system students build.
Wheel encoders + dead reckoning to navigate precise distances and angles. Students fuse encoder data with IMU readings using a Kalman filter — same technique in real autonomous vehicles.
Ultrasonic ranging + state machine logic. Navigates a cluttered room autonomously. Introduces real-time decision loops in Python.
Source a real problem from a real person. Build a working robotics or AI solution. Ship it to that user. No simulations. A real human uses what you built.
Build a robot controlled by decision trees, then upgrade it to a simple neural network. Students see exactly what changes — and why learned models sometimes beat hand-coded rules.
Fuse encoder, IMU, and ultrasonic data through a Kalman filter to produce clean position estimates. The same sensor fusion technique used in drones, autonomous cars, and spacecraft.
Students train a small neural network using TensorFlow Lite on a Raspberry Pi camera. The robot identifies and sorts colored blocks — a real edge-AI inference pipeline built from scratch.
Q-learning agent trained to navigate a simulated grid environment through trial and reward. Students then transfer the learned policy onto a physical robot — bridging simulation and reality.
The engineering portfolio, competition results, and technical depth students build through RoboLab opens doors at the most selective programs in the country.
Among programs where RoboLab alumni have been admitted.
Every AI concept is tied to a physical project. Students don't read about machine learning — they build, train, and deploy it on hardware they made.
I'm a robotics engineer with a Master of Science from the University of Pennsylvania and a Bachelor of Science from UC San Diego. I spent years building autonomous drone systems designed to navigate Martian cave networks — and before that, I co-founded two robotics startups where I shipped real hardware for four years.
I teach because robotics, AI, and entrepreneurship are converging into the future of jobs — and the earlier you build real skills, the further ahead you land. This course is designed to give students the exact foundation I wish I'd had at their age.
No credit card required. No prior experience needed. Just curiosity and a willingness to build real things.
