This is for people who already know the basics aren't enough.
Drribtex brings together practitioners building real robotic systems — not hobbyists, not curious beginners. If you are designing autonomous pipelines or integrating learned models into hardware, this is where the work gets serious.
Curriculum updated quarterly
The material tracks what is actually shipping in labs and production environments. When transformer-based planners displaced older architectures in 2023, the curriculum changed within weeks — not at the next annual revision.
Instructors with active projects
Every instructor is currently working on robotic systems outside of teaching. That means the examples are drawn from real constraints — budget limits, hardware failures, regulatory friction — not textbook scenarios.
students have completed at least one full program since 2017
Async-first, live when it counts
Sessions are recorded and indexed by concept so you can find the exact segment you need. Live reviews happen across time zones — Jerusalem, Singapore, São Paulo, Berlin — and are scheduled around participants, not instructors.
What participation actually requires
Most of the cost here is time, not money. A typical program runs 14 weeks. Participants who get the most from it are spending 8–12 hours a week — not skimming lectures, but running experiments, breaking things, and writing up what they found.
The financial side is structured to reflect what you are getting: access to specific expertise for a defined period, not a subscription to content you may never use.
A selection, not a catalogue
Four programs are running at any given time. Each covers a distinct problem area — they are not sequential, and you do not need to take them in order.
Autonomous Systems Masterclass
From sensor fusion to decision-making under uncertainty. Covers real deployment constraints that simulators do not replicate.
Program details
Perception and Navigation in Robotics
Computer vision pipelines, SLAM variants, and how to choose the right approach for your hardware budget and latency requirements.
Program details
Robot Learning from Demonstration
Imitation learning, reward shaping, and the practical gap between what a model learns in simulation and what it does on the floor.
Program details
Edge AI for Embedded Robotics
Running inference on constrained hardware — quantization, model pruning, and the tradeoffs you will face before a single line of code is written.
Program detailsWhat people said after finishing
"I had read all the papers on imitation learning but kept hitting the same wall when I tried to implement anything on actual hardware. The program forced me to slow down and understand why the gap exists — not just that it does."
"The instructor had deployed the exact architecture we were discussing on a real warehouse system six months earlier. That context changed how I understood every tradeoff. No textbook gives you that."
"It took me longer than I expected to get through the navigation module — about three weeks more than the schedule suggested. But the live review sessions made it possible to keep going without feeling lost."
The difference is in what gets skipped
Plenty of platforms teach robotics through curated demos and clean datasets. The results look good in a browser. They tend to fall apart when the sensor is noisy, the lighting changes, or the motor driver has a 40ms lag that was not in the spec sheet.
Drribtex does not protect participants from those conditions. The exercises are built around them. You will work with hardware that misbehaves, models that fail, and schedules that do not allow for a clean restart.
Founded in Jerusalem — first cohort focused on industrial arm control with early deep learning tools
Expanded to autonomous mobile platforms; added embedded inference track after participant demand
Curriculum restructured around foundation models entering robotic control pipelines
Live review format extended globally; participants now join from 19 countries across 4 continents