AI Engineer
AI Engineer with 2+ years focused on ML/DL. Specializing in NLP, LLMs, and computer vision — with hands-on experience fine-tuning Arabic language models, building face recognition systems, and implementing transformer architectures from scratch.
chapter i
AI Engineer with 2+ years focused on ML/DL. Specializing in NLP, LLMs, and computer vision — with hands-on experience fine-tuning Arabic language models, building face recognition systems, and implementing transformer architectures from scratch.
appendix a
Python, SQL, Java, C/C++, JavaScript, HTML/CSS
PyTorch, Scikit-learn, OpenCV, Hugging Face, Transformers, NLP, Computer Vision
Jupyter, Pandas, Seaborn, Matplotlib
PostgreSQL, MySQL
Docker, Git/GitHub, Hugging Face
folio ii–vi
Face recognition system for employee attendance logging. Fine-tuned on LFW dataset using ArcFace loss head. Achieved 98% ranking. Built with PyTorch + OpenCV.
stack: PyTorch, OpenCV, ArcFace, LFW Dataset
Fine-tuned AraBERT to handle Hadith matn and sanad separation — distinguishing text content from chain of narration.
stack: AraBERT, Hugging Face, NLP, Transformers
Implemented the Transformer architecture from scratch by reading the 'Attention is All You Need' paper — zero boilerplate, full understanding.
stack: PyTorch, Transformers, Attention, Deep Learning
Admin dashboard for Riwaqi e-commerce platform with inventory management and analytics.
stack: Next.js, Tailwind CSS, PostgreSQL, Drizzle
chronology
Began Computer Science studies focusing on C/C++, data structures, algorithms, and problem-solving.
Transitioned to JavaScript development, building web apps while strengthening algorithmic problem-solving on CodeWars.
Deep-dived into ML/DL — read Hands-On Machine Learning, implemented CNN/RNN/LSTM models, studied the Attention Is All You Need paper, built Transformer from scratch, fine-tuned AraBERT, and built a production face recognition system.
colophon
© 2025 Ismail Medjahdi · AI Engineer
Set in Playfair Display, JetBrains Mono, Inter