Thesis @ AI Medical AG · Research Assistant @ University of Neuchâtel
> building intelligent systems

Abdelrahman Faqieh

AI Engineer

LLMs Generative AI Computer Vision Segmentation

Bern, Switzerland

Get in Touch View Projects Download CV
INPUT HIDDEN 1 HIDDEN 2 OUTPUT LLM text CV vision AI core GEN synth SEG med ŷ1 class ŷ2 pred ŷ3 gen NEURAL NETWORK

About Me

I'm an AI Engineer pursuing my Master's in Artificial Intelligence in Medicine at the University of Bern. I'm passionate about AI in all its forms - from large language models and generative AI to computer vision and segmentation. What excites me is solving hard problems with intelligent systems, regardless of the domain.

Most recently, my work has been at the intersection of AI and medicine - developing deep learning architectures for brain lesion segmentation and building vision-language models for histopathology. But I've also built autonomous path planning systems for robotics and designed metadata-conditioned diffusion pipelines for controllable image generation.

I thrive wherever AI meets a meaningful challenge. Whether it's healthcare, robotics, or creative generation, I bring strong engineering fundamentals, research depth, and a drive to ship real systems.

4+
Work Experiences
3+
Research Projects
4
Languages Spoken

Work Experience

Machine Learning Researcher

AI Medical AG - Master's Thesis

Zurich, Switzerland Mar 2026 – Present

Developing and benchmarking state-of-the-art deep learning architectures for brain lesion segmentation in longitudinal MRI scans and integrating the selected architecture into the company's Jazz software to enable automated lesion detection and progression tracking across patient follow-up studies.

Deep Learning Medical Imaging MRI Segmentation

AI Intern

Institute of Tissue Medicine and Pathology

Bern, Switzerland Nov 2025 – Present

Extended an existing vision-language framework by fine-tuning vision and language encoders on histopathology images and textual annotations, aligning their representations via contrastive loss for unsupervised downstream classification.

Vision-Language Models Contrastive Learning Histopathology

Software Engineer (Part-time)

ARIS Space

Zurich, Switzerland Jun 2025 – Dec 2025

Developed a ROS 2 node in Python for 3D trajectory generation and geometric path planning for the Nautilus autonomous underwater glider, computing turn radii, pitch angles, and waypoints.

ROS 2 Python Path Planning Robotics

Biomedical Engineering Intern

Medical Park Goztepe

Istanbul, Turkey Jan 2024 – Feb 2024

Diagnosed and repaired electrical circuits in medical devices to improve functionality and reliability. Monitored equipment performance and identified potential failures to ensure seamless operation.

Medical Devices Circuit Diagnostics Biomedical Engineering

Projects & Research

AI Internship - Institute of Tissue Medicine and Pathology

Vision-Language Contrastive Learning for Histopathology Classification

PHASE 1: TRAINING Single Source Dataset (image-text pairs) Pathology Slide Doctor Annotation Vision Encoder Text Encoder CL Loss PHASE 2: INFERENCE New Slide (unseen disease) "tumor cells" "necrosis" "stroma" "lymphocytes" Class Descriptions Vision Enc. Text Enc. z_img z_t1..z_tN cos(z_img, z_t) argmax → class PHASE 3: GENERALIZATION Trained on 1 dataset → Generalizes to many Public Datasets TCGA, Camelyon... Private Datasets Hospital cohorts Breast cancer Lung tissue Colon polyps Zero-shot accuracy: Dataset A - 80% Dataset B - 71% Dataset C - 85% No labels needed at test time

Fine-tuned vision and language encoders on a single histopathology dataset using contrastive loss to align image-text representations. At inference, a pathology slide passes through the vision encoder and candidate class descriptions through the text encoder - the predicted class is chosen by cosine similarity. This approach generalizes zero-shot to unseen diseases and both public and private datasets.

Code under NDA
Vision-Language Models Contrastive Learning Histopathology Zero-shot Classification Fine-tuning
RESEARCH ASSISTANT - UNIVERSITY OF NEUCHÂTEL

Metadata-Conditioned Diffusion for Synthetic Medical Image Generation

Clinical Metadata (age, diagnosis, site...) PATH A: LANGUAGE CONDITIONING Template Prompt Builder metadata → text Stable Diff + LoRA Generated A PATH B: DIRECT ENCODING Learned Encoder → embeddings Stable Diff + LoRA Generated B Doctor Evaluation Clinical accuracy scoring WHICH METHOD PRESERVES MORE? Language conditioning: Clinical Direct encoding: Clinical Disease domains tested: Skin Cancer Dermatology lesion patterns Heart & Lungs Chest X-ray cardiac anomalies Retina Fundoscopy diabetic detection

Compared two LoRA fine-tuning approaches for medical image synthesis using Stable Diffusion: one uses predefined templates to convert clinical metadata rows into natural language prompts, the other encodes metadata directly into learned embeddings that condition the diffusion process. Generated images across dermatology, chest X-ray, and retinal domains were evaluated by medical professionals to determine which conditioning method best preserves clinically relevant information for downstream diagnosis.

View Code Research ongoing - code in progress
Stable Diffusion LoRA Fine-tuning Conditioning Methods Medical Imaging Clinical Evaluation Hugging Face
Image DINOv2 Encoder Cross-Attn Caption BLEU Evaluation

Image Captioning with Transformers and DINOv2

Developed six progressively improved models to generate captions for images. Used pretrained DINOv2 encoders and evolved decoders from RNNs to Transformers with cross-attention over spatial tokens. Evaluated with BLEU scores and visualized attention maps.

Transformers DINOv2 Computer Vision NLP
Sensor Data Feature Extraction ML Model A+ Quality Real-time Pipeline

WiiFit Exercise Quality Detection With Machine Learning

Designed and implemented a real-time machine learning pipeline for exercise quality assessment using WiiFit sensor data, including feature extraction, model training, and performance evaluation in a team-based setting.

Machine Learning Sensor Data Real-time Signal Processing
Ongoing Research

Distributed Learning Systems Lab - University of Neuchatel

Conducting research on metadata-conditioned diffusion models for controllable image generation, leveraging structured (tabular) data for cross-modal conditioning.

Diffusion Models Cross-modal Research

Technical Skills

Programming Languages

Python C++ C Matlab R

Deep Learning

PyTorch Scikit-Learn Hugging Face LangChain timm

Computer Vision & NLP

OpenCV Open3D SimpleITK spaCy NLTK

Data & C++

Pandas NumPy Matplotlib Seaborn STL Eigen OpenMesh

Tools & Infrastructure

Git Docker Slurm Weights & Biases Metashape Blender

Languages

English (Native) Arabic (Native) Turkish (B2) German (A2)

Education

M.Sc.

Artificial Intelligence in Medicine

University of Bern

Bern, Switzerland · Sep 2024 – Sep 2026

Deep Learning, Machine Learning, Computer Vision, From NLP to LLMs, Modeling & Scaling of Generative AI Systems, Reinforcement Learning, Trustworthy AI, HPC & Cloud Computing, C++ Programming

B.Sc.

Biomedical Engineering

Bahcesehir University

Istanbul, Turkey · Sep 2020 – July 2024

Principles of AI, Linear Algebra, Differential Equations, Signals and Systems, Programming in Python & C, Biostatistics, Medical Imaging & MRI, Modeling and Simulation

Honors Graduate

Honors & Awards

Honors Graduate - Graduated with Honors in Biomedical Engineering from Bahcesehir University
Koc University Summer Academy - Pass with Distinction in Life Sciences Course

Get in Touch

I'm always open to discussing new opportunities, research collaborations, or interesting projects. Feel free to reach out!

Email abdelrahman.faqieh@gmail.com GitHub github.com/afaqieh LinkedIn linkedin.com/in/abdelrahman-faqieh
Location Bern, Switzerland