About

Learn more about me

Elsayed Mohamed

AI & Software Engineering Innovator

Driven by passion and guided by discipline, I'm a Professional Machine Learning Engineer & Researcher with a BSc degree in Electronics & Communications Engineering.

  • Current Degree: BSc. of Electronics & Communications Engineering
  • Email: sayedmohamedscu@gmail.com
  • Consulting & Freelancing: Available

With 5+ years of experience in various startups and as a freelancer, I've tackled a diverse set of challenges. My expertise spans software & hardware, from building apps and designing drones to developing cutting-edge machine learning applications. I'm constantly experimenting with new ideas and technologies, driven by an insatiable curiosity to learn and innovate.

My focus is on developing deep/high tech and building Machine Learning solutions that deliver real value to businesses. I've contributed to multiple groundbreaking projects across various domains:

  • Advanced Medical Imaging @ PaxeraHealth:
    • Architected fast 3D tumor segmentation solutions for kidney, pancreas, and colon
    • Achieved state-of-the-art inference times in 2D and 3D pipelines
    • Implemented 2D segmentation using the SAM model
    • Developed an AutoML pipeline for x-ray and OCT imaging
  • AI-Powered Documentation @ PaxeraHealth:
    • Created a robust FAQ solution using advanced NLP techniques
    • Collaborated with cutting-edge language models (OpenAI, Llama2/3, Mixtral, Phi2/3)
    • Fine-tuned Vision Language Models (VLM) for medical applications
  • Tuba Auto-Annotation AI-Vision Platform @ DevisionX:
    • Led development of auto-labeling tools using Vision Transformer models
    • Optimized AutoML pipeline for image classification and object detection
    • Conducted research on vector quantization VQVAE for interpretable classifiers
  • AI-Powered e-KYC Platform @ Digified:
    • Built machine learning models for reading Egyptian ID data
    • Developed text de-noising and segmentation autoencoder models
  • Access Control Solutions @ PassApp:
    • Engineered pipelines for license plate recognition and face matching
    • Created solutions for Egyptian national ID number recognition

I firmly believe that the fusion of technical innovation and community engagement is at the core of advancing engineering as a discipline. My commitment to both technical excellence and community-driven initiatives is the driving force behind my vision.

Companies

Private Contracts

Years of Experience

Talks & Sessions

Business Domains

FinTech
Telecom
Manufacturing & Industry 4.0
Digital Transformation
Robotics (Drones / UGV / Self Driving )
Access Control

Technical & Research Interests

Machine Learning & Deep Learning

Vision Langauge Models

3D/2D Medical Imaging

MLOps

Digital Signal Processing

Wifi Sensing

Systems Engineering

Access control (LPR /ID & Face Recognition)

GEN-AI

DevOps

Drones and Robtics

Automatic Speech Recognition

Formal Software Design Methods

Mathematical Modeling

Natural language processing

Applied Mathematical Optimization

Applied Category Theory

Applied Artificial Intelligence

Control Theory

Systems Theory

Resume

Check My Resume (Details on LinkedIn)

Professional Experience

Full Time

Senior Computer Vision engineer at PaxeraHealth

November 2023 - Present

Boston / Remote Egypt Office

Computer Vision engineer lead at DevisionX

febroury 2023 - Present

Cairo / Remote

Co-Founder & Computer vision Lead at Pxivision

March 2023 - Present

Cairo / Remote

Computer Vision Consultant at DroneBase

febroury 2023 - May 2023

Italy / Remote

Computer vision Engineer at PassAPP

June 2022 - January 2023

Cairo / Remote

Computer Vision engineer at DevisionX

July 2020 - November 2020

Cairo / Remote

Computer vision Engineer at Digified

July 2019 - November 2019

Cairo / Remote



Fiverr Level1 seller



Education

Bachelor of Engineering in Electronics & Communication Engineering

2016 - 2021

Suez Canal university, Egypt

Thesis : Human Gait recognition and intrusion detection Using Wifi sensing and Deep Learning (Blaze-WI)

Gallery

My Works

  • All
  • App
  • Card
  • Web

App 1

App

Web 3

Web

App 2

App

Card 2

Card

Web 2

Web

App 3

App

Card 1

Card

Card 3

Card

Web 3

Web

AI Models

My Expertise in AI & Machine Learning Models

Computer Vision

Image Classification

  • Transformers (ViT, Dino2)
  • EfficientViT, MobileViT
  • Convolutional Neural Networks (CNNs)

Object Detection

  • YOLO (v3 to v10)
  • DETR model
  • Faster R-CNN, SSD, RetinaNet

Segmentation

  • SAM / SAM 2, EfficientSAM
  • UNET (CNN based)
  • NNU-Net (CNN & ViT based)
  • Mask R-CNN, DeepLab

Advanced Vision Tasks

Pose Estimation

  • OpenPose
  • Google Pose

Face Recognition

  • FaceNet
  • DeepFace
  • ArcFace

Visual Tracking

  • SORT
  • DeepSORT

OCR

  • TrOCR
  • PaddleOCR
  • Tesseract
  • EasyOCR

Vision-Language Models

  • PaLI
  • Florence
  • Phi-3, 3.5
  • LLaVA
  • OpenGPT-4
  • CLIP / SigLIP
  • GroundingDINO

Speech Processing

Speech Recognition

  • Whisper
  • DeepSpeech
  • Wav2Vec

Text-to-Speech

  • Various TTS models

Generative Models

Vision

  • Autoencoder, VAE, VQ-VAE
  • GAN, CycleGAN, StyleGAN
  • Stable Diffusion

Language Models

  • LoRA / QLoRA
  • LLaMA 2 / 3 / 3.1
  • Phi-2, 3, 3.5
  • Mistral

Autoregressive Models

  • PixelCNN

MLOps

Deployment

  • VLLM for production
  • Cloud platforms (AWS, GCP)
  • On-premise solutions
  • ONNX for serving

Optimization & Tracking

  • Model Optimization (Pruning, Quantization, Knowledge Distillation)
  • Experiment Tracking (ClearML, Weights & Biases)
  • Data Version Control (DVC)

Pipelines

  • AutoML pipelines
  • CI/CD for ML

This overview represents my expertise across various AI domains. Each category shows specific models and techniques I'm proficient in.

Contact

Contact Me

My Address

Egypt

Social Profiles

Email Me

sayedmohamedscu@gmail.com