About Me

I’m Ahmed Hassan, a researcher and final-year Computer Science and Engineering student at the Egypt-Japan University of Science and Technology (E-JUST), graduating in March 2026.

My mission is to build generative AI that is not only powerful but also reliable, grounded, and multimodal. This conviction was forged through hands-on experience at the frontier of AI research and application. At Microsoft, I tackled the critical problem of LLM hallucination, while my current work at Anovate.ai involves deploying real-time 3D vision models in demanding industrial environments. These experiences have shown me the vast potential of AI and the essential work needed to bridge the gap between theoretical models and real-world utility.

🧑‍🔬 Research & Internships

My journey is defined by a blend of industry-led challenges and academic research, allowing me to see problems from both a practical and a theoretical lens.

• AI Engineer Intern @ Anovate.ai (2025 – Present) I am currently developing and implementing real-time deep learning models for 3D reconstruction.

• Applied Science Intern @ Microsoft (2025) As part of the MSN News team, I focused on AI-generated content moderation. I engineered solutions for hallucination detection in AI summaries, directly impacting model safety. My contributions led to a measurable increase in hallucination detection recall to 80% while simultaneously reducing the content rejection rate to 40%. My mentor noted my proactive approach as testing multiple hypotheses, providing data-driven arguments, and documenting my findings rigorously.

• Research Intern @ Nile University (2024) I explored the application of LLMs to enhance Arabic education. My work involved curating a novel dataset of over 3,000 questions and fine-tuning the Jais model, which boosted its performance by 10% on educational queries and achieved 93% accuracy across diverse domains.

🔬 My Current Research Focus

I am currently immersed in two research projects that sit at the intersection of generative modeling and computer vision, pushing the boundaries of what’s possible in synthesis and reconstruction.

• Generative 3D Cardiac Reconstruction from MRI Data This project tackles the critical challenge of synthesizing high-fidelity 3D cardiac MRIs, especially in the presence of noise and motion artifacts. I designed a teacher-student Generative Adversarial Network (GAN) framework that achieves 92.57% fidelity against real datasets. To enhance clinical utility, I also developed a U-Net-based denoiser that significantly improves image quality (PSNR by up to 5.68 and SSIM by 0.16). By training on simulated clinical noise, the model robustly removes artifacts in over 95% of test cases, showing a clear path toward aiding surgical planning and diagnosis.

• Rethinking Encoder Architectures for High-Fidelity Virtual Try-On Inspired by methods like Stable-ViTON, I am investigating how to achieve more realistic and fine-grained garment transfer. My hypothesis is that current encoders are a bottleneck for capturing complex textures and deformations. My work focuses on designing and integrating spatially-aware encoding strategy to create more convincing and detailed virtual try-on results.

🎯 Core Research Interests

My experiences have solidified my passion for the grand challenges in AI. I am driven by the intellectual puzzle of how we can build machines that perceive, reason, and create more like humans. I am most excited by:

• Generative & Multimodal AI: Combining modalities like vision, text, and 3D geometry to build holistic models (GANs, VAEs, Diffusion Models).
• LLM Reasoning & Grounding: Moving beyond pattern matching to enable LLMs to reason logically, interact with structured data, and ground their outputs in verifiable facts.
• Multi-Agent Systems: Designing collaborative AI systems where multiple agents can work together to solve complex, multi-step problems.
• Cross-Modal Tasks: Building the bridges that connect computer vision, NLP, and representation learning to solve tasks like text-to-motion synthesis or visual question answering.

At its heart, my work is motivated by the beautiful intersection of creativity, mathematics, and utility.

🌍 Beyond the Lab

Outside of research, I believe in sharpening my mind through different disciplines and hobbies:

• 📚 Reading: I’m always exploring new papers, but I also love books that connect different fields. I’m currently reading Soccermatics to see football not just as a sport, but as a dynamic system of data, probability, and emergent strategy.
• ♟️ Chess: A perfect arena for honing patience, strategic thinking, and the ability to anticipate future moves-skills surprisingly relevant to research planning.
• ⚽ Football: Both playing and analyzing the game keeps me active and appreciative of teamwork and system dynamics.
• 🌐 Competitive Programming: I actively compete on Codeforces to keep my algorithmic problem-solving skills sharp and efficient.
• 🎬 Film Analysis: I enjoy breaking down films to understand their structure, themes, and visual language. You can find my thoughts and ratings on my Letterboxd.

🚀 Looking Ahead: My Research Vision

I am actively applying to M.Sc. and Ph.D. programs to begin the next chapter of my journey. My long-term goal is to become a research scientist dedicated to tackling fundamental challenges in AI. I want to build systems that can understand the world with common-sense, reason across different types of information, and generate content that is not only creative but also truthful and helpful.

For me, the ultimate pursuit isn’t just to build better models; it’s to build understanding - creating systems that can perceive, reason, and create alongside us to accelerate scientific discovery and enrich human life.

📌 Quick Links

If you’d like to explore more about my work, feel free to check out the following sections:

• 📄 Publications – My published and submitted research.
• 📝 CV – A detailed overview of my academic and professional background.
• 🏆 Awards – Recognitions and achievements.
• 👨‍🏫 Mentoring – My experience guiding and mentoring students.
• 💻 Projects – Hands-on projects showcasing applied AI systems.