Mohamed Abdelfattah
🚀 Training Models, Delivering Results

Call me Os

Mohamed Abdelfattah

PhD-ing Computer Vision @ EPFL Switzerland 🇨🇭
Ex-Intern @ Meta Reality Labs
VITA Lab · Luckily advised by Alexandre Alahi
First-author at:
Research Snapshot My research asks a simple question: can machines learn strong representations by predicting the world rather than labeling it? I study multimodal transformers and large language models that integrate perception, motion, and language. By training models to infer missing structure across time and modalities, I develop predictive pretraining methods that yield transferable representations for action recognition, egocentric vision, and world modelling. Multimodal Learning Egocentric Vision Action Recognition World Modelling Self-supervised Learning LLMs For recruiters: I am actively looking for full-time research positions in computer vision. Feel free to reach out.

📚 Publications

💼 Experience

Meta Reality Labs logo

AI Research Scientist Intern

Meta Reality Labs

Zurich, Switzerland
  • Manager: Edoardo Remelli. Advised by: Eric Sauser and Shugao Ma.
  • Built multimodal LLMs for smart wearable systems.
  • Spearheaded the collection of a novel dataset and benchmark for robust human-state perception.
  • Published as first author at CVPR 2026.
SpreeAI logo

Computer Vision Research Scientist

SpreeAI

Nevada, USA
  • Built a conditional diffusion model for high-fidelity face reconstruction.
  • Improved reconstruction performance by 20% over prior methods.
  • Designed a U-Net head-swapping pipeline preserving pose, skin tone, and lighting.
Thuwal, Saudi Arabia
  • Worked with Prof. Mohamed Elhoseiny in KAUST Vision-CAIR.
  • Led collection of a large vision-language dataset with affective captions/explanations in 4 languages.
  • Co-developed training recipes leveraging cultural and linguistic diversity.
  • Co-authored a paper published at EMNLP 2022.
Boston University logo

Remote Visiting Research Student

Boston University

Boston, USA
  • Worked with Prof. Sarah Adel Bargal in the IVC Group.
  • Built data-augmentation pipelines preserving object traits and dataset diversity.
  • Analyzed dataset balancing strategies for segmentation of deformable waste objects.
  • Co-authored a paper published at CVPR 2022.

🎓 Education

Lausanne, Switzerland
  • Supervised by Alexandre Alahi at VITA Lab.
  • Building SOTA frameworks in Multimodal Large Language Models (MLLMs), self-supervised learning, and action understanding.
  • First author of top-tier conference publications at CVPR, NeurIPS, and ECCV; co-author at EMNLP.
The American University in Cairo (AUC) logo

B.S. in Computer Engineering

The American University in Cairo (AUC)

Cairo, Egypt
  • Graduated with Highest Honors (Rank: 5/80, GPA: 3.91/4.0) and recognized on the Dean’s List of Honors.
  • Double Minor in Mathematics and Business Administration.
  • Thesis: Fine-Grained Text-to-Image Generation using Generative Adversarial Networks (GANs).
  • IELTS 8.5/9.0, GRE: Math 170/170, Verbal 161/170, Writing 5.5/6.

🏆 Awards

A curated collection of milestones and recognitions.

2022

Awarded the AUC PA Cup for the class of 2022 for top academic and extracurricular achievements.

AUC

2022

High Academic Achievement (Top 10% of Graduating Class) in SSE Honours Assembly.

AUC

2021

Undergraduate Research Award, 4,000 USD.

AUC

2019

Ranked among the top 10 ROV teams in the Middle East.

MATE ROV Regional Competition

2018

Ranked 1st in the CSCE programming contest.

AUC

2018

Named Highest Achiever and Reader of the Year.

AUC

📖 Teaching

Instructional roles showcased by course focus, not timeline order.

2021 & 2022

Computer Architecture Teaching Assistant

AUC

2022

Deep Learning Teaching Assistant

AUC

2023–present

Head Teaching Assistant (TA) of Deep Learning for Autonomous Vehicles

EPFL

✉️ Get In Touch

Connect

I am always happy to discuss research collaborations, internships, and full-time opportunities in computer vision and multimodal learning. If you are hiring, collaborating, or want to chat about ideas, feel free to reach out.

✓ Copied to clipboard!
Send a message

Find me on