Morph

When I arrived the team was a professor and a PhD student. There was a physical prototype of the first product, but no electronics. I chose the MCU, defined the architecture, wrote the initial firmware, and built out the rationale for why things were done the way they were. That foundation is what the team of 40 now builds on.

Across the project I have designed around 12 unique boards, with over 1,000 units ordered in total. Board sizes range from 100×80mm down to 12mm radius. All have been designed in KiCad.

The more demanding work involved a 6-layer board — the smallest achievable with our manufacturer — with a chip antenna using a coplanar waveguide design. Getting the antenna to propagate outward correctly required careful RF layout and ground pour management. A second antenna design ran into a different problem: the enclosure formed a Faraday cage. Understanding why the antenna was failing, and what to do about it, was a useful lesson in the gap between RF theory and physical installation.

Power design included a system running from an 18650 cell, regulated to 5V at up to 6A, designed to survive the sharp current spikes from motors starting under load. Battery charging and PMIC integration, ground layer management, and noise reduction have all been recurring concerns.

More recently the work has shifted toward DFM: selecting components that are in production and available in volume, working with contract manufacturers to validate that the designs can be assembled repeatably, improving silkscreen legibility and layout clarity, and identifying noise issues before they become production problems.

A VP of Engineering joined and now oversees the engineering function. My role became more focused on PCB design and assembly, with Zephyr RTOS firmware work alongside.