Looking for a job

Youssef Medhat

@youssefmedhat

Machine Learning Engineer building production GenAI, RAG, forecasting, and CV systems—end-to-end from pipelines to deployment.

Egypt

Message

What I'm looking for

I’m looking for a team where I own the full ML lifecycle and ship production GenAI/ML systems end-to-end—especially secure/offline RAG, predictive analytics, and measurable reliability through monitoring, observability, and iteration.

I’m a Machine Learning Engineer with production experience across GenAI, computer vision, forecasting, and large-scale data platforms. I’ve built end-to-end ML pipelines from data ingestion and feature engineering to training, deployment, monitoring, and optimization, working closely with government and enterprise stakeholders on national-scale AI initiatives.

Some of my proudest work includes a production-grade, CPU-only, airgapped enterprise RAG system for secure document Q&A, plus an AI-powered LMS assistant integrated with Open edX (LTI 1.3) using LangGraph agents, guardrails, and automated content ingestion. In predictive analytics, I led SmartCal (an automated probability calibration framework) and delivered 90%+ post-calibration accuracy, with MLflow deployment, drift detection, and automated retraining; I also built a telecom KPI and big data platform using Kafka into ClickHouse orchestrated via Airflow, and implemented real-time traffic classification with RiverML. Alongside my current M.Sc. in Machine Learning at Georgia Tech, I support research as a Remote Research Assistant at Dilab—designing RL-based personalization/recommendation pipelines—and I’m published in AutoML 2025 with additional papers accepted at AIED 2026 and ITS 2026.

Experience

Work history, roles, and key accomplishments

Current

Research Assistant

Current

Dilab

Jan 2025 - Present (1 year 3 months)

Designed predictive personalization and engagement models, including persona clustering and knowledge-gap detection. Architected an RL-based recommendation pipeline with offline policy evaluation for teacher-facing insights (paper submitted).

Knowledge Gap Detection Behavioral Modeling RL Based Recommendation Predictive Analytics

Current

Machine Learning Engineer

Current

Giza Systems

Aug 2023 - Present (2 years 8 months)

Designed and delivered production-grade AI/ML systems across GenAI RAG, computer vision, forecasting, and large-scale data platforms. Built a secure CPU-only airgapped RAG for the Ministry of Defense and delivered SmartCal probability calibration achieving 90%+ post-calibration accuracy, plus real-time traffic classification via Kafka→ClickHouse pipelines.

RAG LangGraph MLFlow Airflow Kafka Clickhouse Elasticsearch Ollama Minio

Frontend Angular Developer

Rubix Marketing Solutions

May 2023 - Aug 2023 (3 months)

Built dynamic Angular-based web applications by translating design mockups into responsive UI.

Angular Frontend Development Responsive UI Web Applications

Data Science Intern

Egyptian Space Agency

Jul 2021 - Sep 2021 (2 months)

Applied machine learning techniques to satellite telemetry data to develop anomaly detection and forecasting models.

Anomaly Detection Forecasting data preprocessing Model Evaluation

Education

Degrees, certifications, and relevant coursework

Georgia Institute of Technology

Master of Science, Computer Science (Machine Learning)

2024 - 2026

Grade: 3.71

Activities and societies: Research: Personalization Pipeline with RL-based recommendation engine (Dilab); multi-agent DRL coursework (Overcooked environment).

Completing an M.Sc. in Computer Science (Machine Learning), including coursework and research on multi-agent deep reinforcement learning and an RL-based personalization/recommendation pipeline.

Helwan University

Bachelor of Science, Computer Science

Grade: GPA: 3.28/4.0

Activities and societies: Minor: Information Systems; project work: Skin Cancer Classification (Ensemble Deep Learning, Grade: A+), ML workflow on AWS SageMaker.

Earned a B.Sc. in Computer Science with coursework and projects including ensemble deep learning for skin cancer classification.