3-DOF Robotic Arm Visualizer
Drag the gripper to move the end effector. Drag elsewhere to orbit.
In my BSc thesis I designed and built a 3-degree-of-freedom robotic arm end to end, from the mechanical geometry through servo actuation, the forward and inverse kinematics in the controller firmware, and the driver PCB. Seven years later I rebuilt the kinematics in the browser as an interactive WebGL model: drag the end effector and the three joint angles solve in real time, move each joint slider and the arm pose updates instantly. React Three Fiber for the scene graph, custom math for the kinematics, no external physics engine.
Two things sit inside this build that a CV cannot carry on its own. The inverse kinematics solver is a few hundred lines of linear algebra I wrote from scratch, not copied off Stack Overflow, so the mathematical layer under the UI is visible alongside the UI itself. Translating the arm from microcontroller firmware into WebGL kept the kinematics faithful between the two mediums, so the system's mental model survived the port. Beyond that, having a working 3D interface in the portfolio is evidence that spatial UIs, GIS layers, CAD surfaces, mission-control panels and operator workflows are inside my range when a role touches them.
Most portfolios show polished React components. This one shows the layer where visual engineering meets physics. The arm is a conversation starter that ties my current frontend work to the industrial-control background I came from, and it is a working proof that the gap between those two worlds is bridgeable. If you are a hiring manager wondering whether the engineer in front of you can learn 3D rendering, GIS visualization, or hardware-adjacent operator UIs on the job, the arm is the evidence that the ramp is short.