Mechanical engineer working in robotics and embedded systems — mostly sensing, control, and the physical hardware that has to make both work together.
I finished a First-Class BEng (Hons) in Mechanical Engineering, with most of what I know about software being picked up when building things that needed it, rather than the other way round. My background is heavier on hardware and physical systems; three years machining, laying up composites and designing drivetrain components for a human-powered vehicle project; hence why I lean towards projects with tangible components rather than pure software work.
My recent work has focused on combining biological sensing, computer vision and embedded robotics - from a real-time EEG-controlled mobile robot, to synchronised multimodal neural-motor recording and edge perception systems. Examples of this are:
EEG-controlled mobile robot. A real-time pipeline which reads EEG from an Emotiv Insight headset to drive a robot over a custom radio link (Arduino + nRF24L01+). I created and integrated: packet-sequencing with duplicate-packet rejection, for an unreliable RF link); a watchdog, assuming some transmissions may vanish mid-command, with failsafe stopping (which triggers before you'd notice something's wrong) and performed an automatic reconnection/reboot procedure, depending on signal-loss severity.
C/C++ Arduino NRF24L01+
Multimodal neural-motor sensing. I'm extending that into a synchronised acquisition pipeline, with, EEG, event markers and vision-derived kinematics combined into one LSL/XDF recording, analysed with MNE. The goal isn't a fixed EEG command classifier; it's letting the system establish subject-specific baselines, to figure out which features are actually predictive, rather than assuming which cortical regions/bands should matter individually.
Python LSL XDF MNE
multimodal-sensing
Edge perception. Pose estimation and computer vision running on an NVIDIA Jetson, feeding into the same multimodal pipeline above (i.e. vision-derived kinematics).
Python NVIDIA Jetson Computer Vision
Biosignal acquisition hardware. In progress: custom EEG acquisition hardware built around the ADS1299, with an STM32 front end. For my PCB design work I have been using KiCad — moving the project outside of consumer headsets, into fully custom, self-built hardware.
ADS1299 STM32 KiCad
- First-Class BEng (Hons) Mechanical Engineering
- Robotics and mechatronic system development
- Embedded C/C++ and Python
- Sensor, actuator, and communications integration
- CAD, mechanical design, and rapid prototyping
- Custom electronics and PCB design
Programming: C/C++, Python, MATLAB
Embedded: Arduino, STM32, NVIDIA Jetson, ESP32
Sensing: LSL, XDF, MNE, EEG, IMU, computer vision
Hardware: KiCad, PCB design, RF communications, motor control
Mechanical: SolidWorks, Fusion 360, ANSYS, CNC, additive manufacturing