Skip to main content
Portrait illustration of Itamar Tsayag

Itamar Tsayag
Senior Algorithm Developer · ML & CV consultant

I build computer vision systems that make it to production.

10+ years turning deep learning research into shipped products in medical imaging, biometrics and 3D vision, at Viz.ai, AiVF and Magic Leap, and today as a hands-on consultant for startups.

years in CV & ML
10+
companies & clients
8
published case studies
7

Companies I've built with

  • Viz.ai
  • Magic Leap
  • AiVF
  • AgileMRI
  • Sanas AI
  • GovPlus
  • Predev
  • 121

What I do

Where I can help

  • 01

    Medical imaging

    A CT perfusion stroke algorithm deployed in 1,500+ US hospitals, IVF embryo grading in clinical use, MRI reconstruction and automated cognitive screening.

    Viz.ai · AiVF · AgileMRI

  • 02

    Faces & biometrics

    Facial landmark detection, geometric registration and ICAO-compliant passport photo validation running at scale.

    GovPlus

  • 03

    3D & spatial vision

    Stereo depth estimation trained on synthetic 3D oral scans, and ToF sensor calibration for augmented reality hardware.

    Magic Leap · Mantis Health

  • 04

    Efficient, real-time models

    Sparse networks found at initialization, real-time speech model inference, and thermal obstacle detection for low-altitude navigation.

    MSc thesis · Sanas AI

Selected work

Case studies

Cover figure for Automated Clock Drawing Test Scoring System

Automated Clock Drawing Test Scoring System

Deep learning pipeline combining YOLOv8, UNet segmentation, and geometric analysis for automated cognitive assessment. Produces standardized 0-10 point CDT scores from hand-drawn clock images for dementia screening.

  • Medical AI
  • YOLOv8
  • UNet
  • +4
Cover figure for AI-Powered Embryo Selection for IVF

AI-Powered Embryo Selection for IVF

Deep learning-based automated embryo assessment system integrating segmentation, classification, and temporal reasoning. Multi-modal framework combining U-Net, ResNet, Video Swin Transformers, and Siamese networks for comprehensive embryo evaluation from time-lapse microscopy.

  • Medical AI
  • Deep Learning
  • Video Classification
  • +5
Cover figure for AI-Driven Passport Photo Validation

AI-Driven Passport Photo Validation

Multi-stage passport photo validation pipeline for ICAO compliance. 98-point landmark detection, geometric registration, and automated cropping at scale.

  • Computer Vision
  • Facial Landmark Detection
  • Deep Learning
  • +1
Cover figure for AI-Powered CT Perfusion Analysis for Stroke

AI-Powered CT Perfusion Analysis for Stroke

Automated CT perfusion analysis pipeline for acute stroke. Core and penumbra tissue classification in under 90 seconds to support thrombectomy decisions.

  • Medical Imaging
  • Stroke Analysis
  • Deep Learning
  • +2
All 7 projects

What clients say

“Itamar combines strong technical skills with a practical, results-oriented mindset. He doesn't just build models—he thinks about how to make them sustainable and usable.”

David Corcos

Founder & CEO, Agile MRI

FAQ

Common questions

Who is Itamar Tsayag?

Itamar Tsayag is a senior computer vision algorithm developer and hands-on ML consultant based in Givatayim, Israel. He has 10+ years of experience building deep learning systems for medical imaging, biometrics and 3D vision, including roles at Viz.ai, AiVF and Magic Leap.

What kind of computer vision work does Itamar do?

He works across the full path from research to production: medical imaging (stroke CT perfusion, IVF embryo grading, MRI reconstruction), faces and biometrics (landmark detection, passport photo validation), 3D and spatial vision, and efficient real-time models such as sparse networks and fast inference.

Is Itamar available for consulting?

Yes. Since October 2024 Itamar has worked as a hands-on ML and computer vision consultant through PreDev Consulting, helping startups and research teams with data strategy, prototyping, model development and production inference. Use the contact page to describe your project.

Does Itamar work remotely with teams outside Israel?

Yes. Itamar is based in Israel and works remotely with teams worldwide. His past work includes a stroke analysis algorithm deployed in more than 1,500 US hospitals and hardware calibration for Magic Leap, a US augmented reality company.

Which tools and frameworks does Itamar use?

Mainly Python with PyTorch and PyTorch Lightning, plus TensorFlow, OpenCV, Docker, MLflow and ClearML. Model choices depend on the problem, from U-Net and ResNet to YOLO, Swin Transformers and Siamese networks.

Have a hard vision problem?

I help teams scope, prototype and ship computer vision and deep learning features, from data strategy to production inference.