Hi, I'm Ahmed 👋
Computer Engineering student building full-stack apps, machine learning systems, and low-level software.
I graduated with a bilingual sign-language recognition platform as my capstone, and I build everything from Flutter apps and REST APIs to 3D medical-image segmentation models and interrupt-driven embedded firmware.
Featured — Graduation Project
Eshara — Bilingual Sign Language Recognition System
B.Sc. Graduation Project · AASTMT, Computer Engineering · July 2026
A real-time, bilingual (American Sign Language + Arabic Sign Language) web platform that recognizes letters and words from a standard webcam and turns them into text and chat messages. Built as a three-tier system — React frontend, Node.js/Express application layer, FastAPI inference layer — powering four independently-trained deep learning models.
0%
ASL Letters
0%
ArSL Letters
0% Top-1
ASL Words
0%
ArSL Words
Four models, four vocabularies
| Model | Architecture | Accuracy |
|---|---|---|
| ASL Letters | MLP on hand landmarks | 99.06% |
| ArSL Letters | MLP on hand landmarks | 99.63% |
| ASL Words | Inception I3D (100 WLASL glosses) | 65.89% Top-1 |
| ArSL Words | Stacked BiLSTM (100 KArSL classes) | 99.62% |
Highlights
- Real-time recognition across 4 modes: ASL/ArSL × Letters/Words
- JWT-authenticated live chat where predictions become messages
- Education Module: sign viewer, sentence builder, guided practice with feedback
- Privacy-preserving inference — only compact landmark vectors leave the browser, not raw video
Projects
Selected work
Skills
Tools I work with
Languages
Frontend & Mobile
Backend
Machine Learning
Data & Infra
Systems
Contact
Let's build something together
I'm open to internships, junior roles, and collaborations across full-stack, ML, and systems engineering. Reach out any time.