Abdelrahman
Mohamed Ismail
I design and ship end-to-end AI systems — from data pipelines and model training to production-ready APIs — across forecasting, computer vision, and large language model applications.
Building systems that turn data into decisions.
I'm an AI Engineer focused on turning research-grade ideas into systems people can actually rely on — forecasting models, retrieval-augmented assistants, and computer vision pipelines built to run in production, not just in a notebook.
My work spans financial time-series forecasting, Retrieval-Augmented Generation, computer vision, and full-stack AI applications. I care about the full lifecycle: clean data pipelines, rigorous evaluation, and APIs that are ready to be shipped rather than demoed once.
Currently completing my B.Sc. in Computer Science and Information Technology — AI & Data Science at the Egypt-Japan University of Science and Technology (E-JUST), and building a portfolio of production-grade systems along the way.
Featured Projects
A selection of AI systems and applications — from financial forecasting to computer vision and full-stack mobile.
An end-to-end forecasting platform for S&P 500 and Nasdaq-100 equities, combining classical ML, deep learning, and financial news sentiment. Benchmarked across 10+ architectures — including LSTM, GRU, N-BEATS, N-HiTS, TFT, Informer, PatchTST, Autoformer, FEDformer, and Mamba — using leakage-free walk-forward validation, with a production-ready RAG layer that generates citation-grounded explanations for every prediction.
A cross-platform smart residential compound management app with role-based dashboards for residents, security guards, and administrators — including QR-based visitor access, real-time gate logging, push notifications, and full audit logging.
A deep learning pipeline for medical image analysis, detecting and classifying brain tumors from MRI scans with high diagnostic precision.
An intelligent assistant combining facial recognition, hand-gesture control, OCR, and voice interaction into a single natural human-computer interface.
A retrieval-augmented chatbot enabling law students to query constitutional documents in natural language, using semantic search and vector indexing over the full legal text.
Translates natural-language Arabic queries into executable SQL using schema-aware prompting and RAG retrieval, optimized for low-latency generation across multi-table schemas.
A decentralized supply-chain tracking ecosystem built on smart contracts, enabling verifiable, tamper-proof tracking of goods across the chain.
Skills & Tools
Languages
PythonC++C#JavaScript / TSDartSQLML & Deep Learning
PyTorchTensorFlowKerasScikit-learnNumPy / PandasLLMs, NLP & CV
RAGLangChainLlamaIndexHugging FaceFinBERTOpenCVMediaPipeOCRBackend, Frontend & Data
FastAPIFlaskREST APIsStreamlitFlutterReactPostgreSQLMySQLMongoDBQdrantChromaDBSupabaseCloud & DevOps
DockerGCPLinuxGitHub ActionsGitPostmanFocus Areas
Time-Series ForecastingRAG SystemsComputer VisionFull-Stack AI AppsExperience & Training
Worked within the Data & Digital Transformation department on enterprise data analytics, reporting workflows, and operational data analysis for a large telecom operator, supporting dashboarding and data-driven decision-making across enterprise teams.
Final-year student building a portfolio of production-grade AI systems alongside coursework in machine learning, deep learning, and data science.
Digital Egypt Pioneers Initiative (DEPI)
Generative AI Professional Track · 200+ HoursBuilt and deployed LLM applications, RAG systems, NLP pipelines, and generative AI workflows.
National Telecommunication Institute (NTI)
Data Science & Machine Learning · 120+ HoursDeveloped machine learning and deep learning solutions using TensorFlow, OpenCV, and Python.
More Projects & Experiments
Beyond what's featured here, I've built numerous AI, machine learning, computer vision, NLP, and full-stack experiments. Explore the full body of work on GitHub.
Let's build intelligent solutions together.
Open to AI/ML internships, full-time roles, freelance projects, and research collaboration.