About
I enjoy the gap between understanding how something should work and actually making it work.
My work sits between math, software, and engineering. I've worked on vehicle optimization at TUfast and built simulation and motion-planning tools. I like turning an idea into something that runs and seeing how it holds up against real-world constraints.
- Studying
- B.Sc. Informatics, TU Munich
- Focus
- Optimization, applied AI, backend
- Languages
- German, English, Arabic, French
Experience & education
-
2024 – 2025
Driving Strategy Engineer
TUfast Eco Team, Technical University of Munich
- Developed the race strategy optimizer for the lux025 solar car (Bridgestone World Solar Challenge 2025), planning speed profiles over 3,000 km under energy and race constraints.
- Extended the PyTorch vehicle model to optimize speed profiles with gradient descent and integrated multiple optimization solvers (IPOPT, ParOpt, ALPSO).
- Built a PySide6 GUI for running optimizations and wrote unit tests for the control-stop simulation (pytest, GitLab CI).
-
2023 – present
B.Sc. Informatics
Technical University of Munich
Coursework spanning data structures, computer architecture, autonomous systems, and numerical analysis.
-
2019 – 2023
Scientific Baccalaureate, Mathematics
Lycée Hrairia 2, Tunis. Highest honors
Projects
Things I've built and shipped on my own.
Cache Simulator
Cycle-accurate cache model
A SystemC simulator comparing direct-mapped and fully-associative caches, with LRU eviction and per-access gate-count tracking.
wifi-to-qr
Command-line utility
Detects the active Wi-Fi connection and prints a scannable QR code in the terminal so others can join instantly.
University projects
Research and coursework at TU Munich.
-
SS 2026
Python for Data Analytics
Data Mining Seminar, TU Munich
A survey of Python's scientific ecosystem (pandas, NumPy, SciPy, scikit-learn) and how it composes into one workflow from data preparation to modeling and evaluation, with a runnable Dash dashboard showing the full pipeline.
-
WS 2025/26
Motion Planning for Autonomous Vehicles
Chair of Robotics & Real-Time Systems, TU Munich
- Converted drone-recorded traffic data into CommonRoad simulation scenarios for benchmarking autonomous vehicle motion planners, using TPH-YOLOv5 for detection and MobileSAM for segmentation.
- Built a validation tool with spatial indexing to detect collisions, off-road driving, and physically infeasible motion, with visualizations for debugging.
-
WS 2023/24 – SS 2024
Computer Architecture & Fundamentals of Programming
TU Munich
- Built a cycle-accurate SystemC cache simulator comparing direct-mapped and associative caches, with LRU eviction and per-access gate-count tracking.
- Solved 35+ Java assignments on OOP, recursion, data structures, algorithms, testing, and concurrency.
Skills
- Programming
- Python, Java, C++, C, JavaScript
- Frameworks & libraries
- FastAPI, React, Node.js, PyTorch, NumPy, OpenCV
- Tools & infrastructure
- Linux, Docker, Git, GitLab CI, pytest, SSH, Proxmox, Ollama
- Languages
- German (native), English (C1), Arabic (native), French (C1)