Embedded Software Engineer
I build the software that connects intelligent algorithms with real hardware, from STM32 firmware to autonomous UAVs.
I’m a software developer focused on embedded systems, computer vision, automation, and intelligent machines.
I enjoy working at the intersection of software and hardware, building systems that interact with the physical world. My experience ranges from STM32 and ESP32 firmware development and motor control to Python-based computer vision, AI applications, Raspberry Pi systems, and autonomous UAV software.
I have developed solutions involving real-time image processing, facial recognition, sensor integration, industrial communication protocols, motor control, autonomous systems, and hardware-software integration. I’m comfortable working close to the hardware while also developing the higher-level software required to turn sensor data into reliable system behavior.
I approach projects with a focus on reliability, real-time performance, clean architecture, and practical problem solving. Whether it is controlling a motor, processing camera data, communicating with an industrial device, or building an autonomous system, I enjoy turning complex technical requirements into working software.
Focus areas
I believe good software should not only work; it should be reliable, understandable, and built to perform in the real world.
I start by understanding the problem, then design a practical, maintainable solution, with clean architecture, systematic debugging and thorough testing wherever software meets hardware. Whether it’s sensors, motors, protocols or computer vision, the goal stays the same: systems that behave predictably, and a little better with every project.
Grouped the way a system is built: the real-time software at its core, how it sees, how it decides and moves, and the tooling around it.
Each project lists the platform, the pipeline and the decisions that mattered. Open one to see how the system is put together.
Featured · Fixed-wing UAV
Autonomous flight and real-time computer vision for a fixed-wing UAV: OpenCV detects and classifies ground targets by colour and shape, and DroneKit flies the mission around what it sees.
14 contributions in the last year · 3 public repos
A practical five-step process I use to turn hardware and software requirements into reliable, tested systems.
We start by understanding the project requirements, hardware, communication interfaces, operating environment, and expected outcome. I identify technical constraints, potential risks, and dependencies before development begins.
I define the system architecture and the interaction between firmware, sensors, motors, communication interfaces, and higher-level software. The goal is to establish a clear and maintainable structure before implementation.
I build an initial working version and test it on the actual target hardware as early as possible. This helps validate hardware integration, communication, control logic, and system behavior before investing time in optimization.
The system is tested under realistic conditions, with each component verified individually and as part of the complete system. I profile performance, identify bottlenecks, fix issues, and refine the implementation through iterative testing.
The final system is delivered with clean source code, configuration, documentation, and the necessary technical details for future development. I aim to leave behind a system that is understandable, maintainable, and ready for further improvements.
Three areas where I take a project from requirements to a working, tested system.
For applications that need to understand and respond to visual information
I develop real-time computer vision solutions using Python, OpenCV, and InsightFace, including face recognition, object and target detection, image processing, and vision-based automation.
Outcome
A working vision pipeline integrated into the target system, with reliable detection and practical real-time performance.
For systems that need to make decisions based on sensor or visual data
I develop software that combines computer vision, sensor data, control logic, and autonomous decision-making, including UAV and other intelligent embedded applications.
Outcome
A functional system that can sense its environment, process information, and execute predefined actions with minimal manual intervention.
For projects where software needs to communicate directly with sensors, motors, controllers, or industrial devices
I work across the hardware and software boundary, integrating components through UART, RS485, Modbus RTU, Bluetooth, and other serial interfaces.
Outcome
A complete hardware-software system where individual components communicate reliably and work together as intended.
Have something else in mind?
Tell me about the project and your timeline, and we'll find the simplest way to work together.
I work on embedded systems, firmware, computer vision, AI applications, automation, motor control, and autonomous systems. I can work on both new projects and existing systems that need debugging, integration, or further development.
Yes. I can review an existing hardware, firmware, or software project, track down firmware, communication, sensor, or motor-control problems systematically, and make the necessary changes without rebuilding the system from scratch.
I have hands-on experience with STM32, ESP32, and Raspberry Pi, as well as embedded Linux environments, and with UART, RS485, Modbus RTU, and Bluetooth communication. I develop firmware, integrate sensors, motors, and peripherals, and build software that talks directly to hardware.
Yes. I can work on anything from a specific firmware fix or hardware integration task to designing and developing a complete embedded or automation system from the ground up.
Ideally, I need a description of the requirements, available hardware, existing code or documentation, communication interfaces, and the expected outcome. If some information is missing, I can help identify the requirements and define the technical approach before development begins.
Tell me in a few sentences what you want to build, what hardware you have and your timeline. Freelance projects, contract work and full-time roles are all welcome.
