How to become a robotics engineer in 2027: a roadmap built from job postings
Choose a robotics role, build the skills it asks for, and show the work. A practical roadmap with checked job examples, project ideas, and posted pay.
Building · robotics, AI, machine learning
Building and explaining intelligent systems.
I build interactive projects and write about robotics, AI, and machine learning. Explore the experiments, the code, and the ideas behind them.
Explore the projects
Start here
Considering a move into robotics, or curious about how this site is built? These guides are a place to begin.
Building a content pipeline that fails early
How this site validates MDX metadata, references, and publication state before a page is deployed.
Learn by changing something.
Explore dot products, conditional probability, and gradient descent through interactive examples. Each lesson includes worked problems and exercises with solutions.
Projects
All projectsPalletizing cell
An interactive isometric robot cell that picks each case off a conveyor, scans the stack and builds an interlocked pallet, layer by layer.
Navier-Stokes visual explainer
Follow a shrinking swirl through five interactive chapters to see how speed can rise while core energy falls.
Portfolio content system
A file-based publishing foundation for project case studies and technical writing on this site.
Writing
All writingBehavior cloning covariate shift: when actions change the inputs
Follow behavior cloning covariate shift through a frozen policy and a two-state motion model. Compare held-out expert actions with learner-state errors, inspect visited states, and test a policy class that can represent the expert.
Behavior cloning from observation: infer missing actions
Learn an inverse model from the learner’s own actions, infer missing expert labels, and test a cloned policy. See how poor exploration and sensing errors change the result.
Controllability and observability: move and reveal state
Find which states an actuator can reach and which a sensor can reveal. Use exact cart motion, two held commands, matrix rank, and indistinguishable initial states.
DAgger imitation learning: label the states your policy visits
Run DAgger imitation learning one round at a time. Collect expert labels on learner rollouts, retrain on the growing dataset, and compare equal label budgets.
Domain randomization: break a training shortcut
Train two policies with the same labels and example count, then randomize one nuisance feature. Inspect the learned coefficients and compare frozen familiar and shifted tests to see what domain randomization changes.
How robotic palletizing works
A palletizing cell picks each case off a conveyor and builds a stable pallet. Learn its parts, the motion and sensing behind each placement, where cells fail, and who makes the hardware.
Now
The new palletizing cell follows a case from the conveyor to an interlocked stack. Pause the motion or step through each placement to examine the arm, vacuum gripper, and pallet pattern.
- Status
- Palletizing cell interactive figure is live
- Explore
- Explore the palletizing cell
About
Interested in how intelligent physical systems work. This site is becoming a record of experiments, technical decisions, and explanations about machine perception, learning, and action. The emphasis is on concrete examples and work that can be examined.
- 01
Choose a question worth making concrete.
- 02
Build the smallest experiment that can expose an assumption.
- 03
Examine what happened and document what changed.
- 04
Publish the result, its limits, and the next useful step.