Said Elhadi

Innovative, results-driven Full-Stack Developer with over 3 years of hands-on experience in creating dynamic and responsive web and mobile applications. Published AI researcher specializing in deep learning in the domain of wireless sensing and computer vision. Originally from Algiers, currently living in Shenzhen, China.

Full-stack web appsAI and machine learningComputer visionWi-Fi sensing researchNext.js and ReactModel optimization

Shenzhen, China

Searchable proof: research, products, and software systems.

Demo Abstract: Underground Root Tuber Sensing via a Wi-Fi Mesh Network

Read

ACM SenSys Conference / 2025 / Said Elhadi, Tao Wang, Yang Zhao

Non-invasive Wi-Fi sensing system using CSI data and deep neural networks to reconstruct cross-section images of potato tubers underground.

DNN Models for WiFi CSI-based Underground Image Reconstruction

Read

ACM MobiSys Conference / 2025 / Said Elhadi, Yang Zhao

Comprehensive study evaluating deep learning architectures for mapping wireless signal tensors to high-resolution underground tuber images.

DRIFT: Data-driven RF Tomography via Cross-modal Sensing and Continual Learning

Read

IEEE AVSS Conference / 2025 / Yang Zhao, Tao Wang, Said Elhadi

Cross-modal sensing system with continual learning for robust underground target detection in dynamic environments.

Projects and experience

Full-stack, AI, computer vision, and wireless sensing work.

Resume

Underground Root Tuber Sensing via Wi-Fi Mesh Network

Published in ACM SenSys '25

Published research on non-invasive Wi-Fi sensing system using CSI data and deep neural networks to reconstruct cross-section images of potato tubers underground.

Wi-Fi CSIDeep LearningESP32PyTorchComputer Vision

Model Inference Optimization

Completed

Optimized model inference reducing 10 batch size input from 22s to 7s and model size from 17MB to ~8MB using ONNX, TVM, and PyTorch.

ONNXTVMPyTorchModel Optimization

Chat2D a 2d ai chat interface

Completed

Built a 2d ai chat interface.

GeminiNextJSShadcnLLM

ESP32 WiFi Mesh Network

Completed

Built mesh network with ESP32C3 WiFi nodes based on Multi-Spin v2.0 architecture to collect CSI data and detect objects.

ESP32Wi-Fi MeshIDF-ESP 5.2IoT

Defect Detection Computer Vision Model

Completed

Developed cutting-edge CNN models using TensorFlow and Keras for detecting defects in visual data, enhancing quality control.

TensorFlowKerasCNNComputer Vision

Crypto Mining Rig & Trading Bots

Completed

Built custom computer for cryptocurrency mining and developed Python trading bots for automated profit trading.

PythonCryptocurrencyHardwareTrading Algorithms