Abdelkareem Elkhateb
Arabic NLP Researcher & AI Engineer
I build efficient AI models, semantic search systems, and production ML pipelines for Arabic and multilingual applications.
Qabilah 
LinkedIn 
X اللغة ليست عِلمًا .. بل هي شيء فوق العلم
لغتنا في خطر داهم .. ونحن أيضًا “Language is not a science — it is something above science. Our language is in danger — and so are we.”
Who am I
I’m an NLP Engineer at Xbites, building the AI backend for Darin.
I also research with Hamza Salem Lab on Arabic embeddings and contribute to NAMMA for open-source Arabic AI.
My thesis focuses on making transformers run efficiently on CPUs. I write about Arabic NLP, embedding models, and building production AI systems.
Projects
BertHash-Femto
113× smaller than AraBERTv2, 94% accuracy. Runs on edge devices. → GitHub
Zarra & Bojji
Tiny Arabic models for phones. → Read more
GPUVec
GPU pricing & ML benchmarks. → Check it
SEO Rat
Free, open-source SEO tool for static sites. Uses real Google Search Console data, runs locally. → GitHub
Recent Writing
My Journey Building a PC for AI Research in Egypt
A personal journey of building a high-performance PC for AI research in Egypt, covering hardware choices, cost comparisons, and local challenges.
LiteRT & Qualcomm AI Hub: On-Device ML Without the Cloud
A practical guide on converting PyTorch models to LiteRT, benchmarking on real Android devices, and using Qualcomm AI Hub for NPU compilation.
Tarteel & Muaalem El Quran: Optimizing Quranic AI
Optimizing the Muaalem El Quran model using TensorRT and ONNX — a deep dive into Quranic speech recognition.
HyperRun + ColGrep: A Self-Hosted Alternative to RunLLM
Add an AI-powered “Ask AI” chat widget to any docs site using ColGrep semantic code search and FastHTML. Self-hosted, BYOK, multi-provider.
View all posts → · Embedding series →
فقالَ: إن تَصدُقِ اللَّهَ يَصدُقْكَ