Hoàng Nam User

At COMBROS Technology Co., Ltd. (Suga Group), I programmed ESP32 firmware for closed-loop PID control, hardware PWM mixing, and Bluetooth SPP decoding to actuate four BLDC motors.

I also trained and quantized a YOLOv8n model on the VisDrone dataset, reducing its file size to 5.8 MB for real-time vehicle and pedestrian detection at over 30 FPS.

On my adaptive hybrid quantum-classical CNN project, I combined six-qubit Variational Quantum Circuits with ResNet18 for medical image classification under data scarcity. The model achieved a 0.919 test ROC-AUC using 10% of training data, and the work was accepted as a poster at the 3rd Youth Science Conference.

I also developed a YOLOv8n road-damage detection pipeline and optimized it for CPU-based edge deployment, reaching 23 FPS with the TensorFlow Lite XNNPACK delegate. I’m a Computer Engineering student seeking a remote AI, computer vision, or embedded role.